In France, call centers occupy a central place in customer relations, with around 3,500 sites spread throughout the country. True showcases for the company, they make it possible to respond to customer requests. In 2023, 78% of French people had contacted customer service at least once, and 55% had done so by telephone, which remains the preferred channel for this type of request¹.

However, the average wait often exceeds their expectations. 45% of customers have a negative memory of their interaction with customer service², and 75.5% say they have switched brands or companies after a bad experience. In this context, while dissatisfaction is mainly due to response times, how can call center practices be adjusted to achieve the long-awaited zero wait time?

Instantaneity and personalization: polarized customer expectations

In a world where everything is accelerating, consumers demand customer service that combines speed and personalization. For 96.5% of them, getting an immediate response is a priority. However, the reality of waiting times remains a major source of frustration: 80% of customers wait more than six minutes before they are taken care of³, while 60% would like assistance in less than a minute⁴. This gap between expectations and reality is widening in the face of increasing call flows and the growing complexity of the cases to be handled. The advisors often struggle to answer all the calls, with an average of 2.8 attempts needed to get an answer⁵.

This tension between the demand for immediacy and the need for a human connection based on empathy and expertise puts the call centers and their teams under pressure. Customer tolerance for waiting is limited: a third of them start to get annoyed after 20 minutes of waiting, and 64% say they would no longer contact the company if the wait exceeded 30 minutes. Poor expectation management can quickly result in a loss of loyalty and a deterioration of the brand image.

To meet these high standards, companies must rethink their approach by integrating tools and strategies that not only reduce waiting times, but also preserve the quality and personalization of interactions.

Zero waiting time: when technology transforms the customer experience

The integration of new technologies is transforming the customer experience. Connected to CRM (Customer Relationship Management) tools, the implementation of chatbots has already made it possible to reduce some of the telephone traffic and provide personalized responses. However, the voice part remains sub-optimized in many companies. Despite the use of Interactive Voice Servers (IVS) for several years, call centers often remain overloaded, resulting in long waiting times, frequent failures and growing frustration among customers.

The integration of callbots in call centers is a real opportunity to improve operational efficiency and customer satisfaction. Available by telephone 24/7, they handle customer calls automatically, in the context of fluid interactions in natural language. They are able to automate certain processes 100% and perform simple first-level tasks such as checking bank overdrafts, providing store addresses and opening hours, etc. They can pre-qualify customer requests, handle calls outside business hours, and make outgoing calls (callbacks, appointment scheduling). By drastically reducing wait times, these tools enable an average productivity gain estimated at 2 hours 20 minutes per day for processing requests.

Companies that adopt these technologies see positive effects: a 20% increase in customer satisfaction⁷, as well as a 40% reduction in the time spent on repetitive tasks⁸. This automation also benefits advisors, whose role is upgraded, strengthening their professional commitment. Freed from these tedious tasks, they can focus on more complex cases that require listening, empathy and personalization, thus improving the customer experience and satisfaction. 60% of employees say they feel more involved in their work when automation frees them from monotonous tasks⁹.

This hybrid model, combining new technologies, artificial intelligence and human intervention, not only relieves the pressure on call centers, but also absorbs requests outside opening hours. It offers a complementary approach where advisors can focus on strategic and rewarding interactions, and free up time for high value-added tasks.

The global customer service market, valued at $16.91 billion in 2023, is expected to grow steadily to $52.54 billion by 2030. With this in mind, call centers need to be transformed to cope with the increase in incoming calls and growing consumer expectations. These centers must evolve gradually to offer a more fluid, fast and customized customer experience.

Sources:

1. Customer Services Observatory of the Year / BVA – 2023
2. Kiamo/Odoxa Study 2023
3. “Vite ma Hotline” Study, En-contact.com
4. Customer Services Observatory of the Year / BVA – 2023
5. Customer Services Observatory of the Year / BVA – 2023
6. An IVR, or interactive voice response, is a computer system that handles incoming calls using recorded voice messages or voice synthesis.
7. Gartner
8. McKinsey study
9. Deloitte

The tourism industry has entered a new era thanks to technological advances, particularly with the integration of generative artificial intelligence. These AI-powered tools are transforming travel experiences and meeting the increasingly demanding expectations of tourists, before, during and after their stays. Digitization has become indispensable for all players in the sector, whatever their field of activity or size. From finding hotel reviews to the best itineraries and activities, to comparing rates, artificial intelligence is involved at every stage of the customer journey. 73% of travelers say they use AI tools for travel-related tasks, a proportion that rises to 87% among 18-24 year olds. Today, it benefits travel agencies and hotel groups looking to improve their targeting. Optimizing operations, forecasting demand, filling: AI accompanies companies on all fronts. It can even help them with predictive analysis, anticipating both the risk of cancellation and market trends. While AI does not generate demand, it is necessary to seize opportunities when they arise. In this context, what are travelers’ expectations?

A frictionless experience and flexible booking management

Organizing a trip can be a time-consuming and complex task, especially as 60% of French people don’t have a specific destination in mind when they start booking their trip. For inspiration, they spend time online, on blogs, social networks, airline websites or watching advertisements. To encourage them to take the plunge, travelers want a seamless experience on online sites, from booking to check-in to itinerary management. The integration of artificial intelligence simplifies this process, offering intuitive navigation and automating numerous tasks such as online check-in or travel document management. In their quest for flexibility when planning their vacations, 41% of travellers want to be able to cancel their reservation free of charge, and 40% want to be able to modify it free of charge. Thanks to advanced technologies such as chatbots and AI systems, capable of handling these requests quickly and efficiently, e-tourism players need to offer flexible booking policies to build trust and customer loyalty in an increasingly competitive market.

Real-time information

Reactivity is a key expectation of travelers. 54% of French people want to be able to book their vacations at the last minute, leading to peaks in connections, bookings and conversations with advisors as the vacations approach. To meet these fluctuations, travel companies need to find new solutions. The integration of AI and, more specifically, chatbots and callbots on websites makes it possible to classify requests, handle those that don’t require the intervention of an advisor, and make others wait for a connection. In the tourism sector, a virtual assistant can respond 24/7 to a variety of requests, such as general information and logistics (check-in, services, rates and conditions, loyalty programs), reservations and purchases (bookings, modifications or cancellations of accommodation, car rentals, plane or train tickets), after-sales service (follow-up information, reservation modifications or cancellations, refund requests), as well as opinions and feedback (satisfaction surveys, suggestions). This permanent availability reassures passengers and enhances their overall experience.

Increased personalization for memorable experiences

Today’s travelers are no longer simply looking for exotic destinations or luxurious hotels. When customers enjoy a personalized experience, they are 44% more likely to become repeat buyers. They look for vacations that provide unique and unforgettable experiences, and prioritize activities and accommodations that allow them to discover local culture, gastronomy or traditions tailored to their specific preferences and needs.

Bots, thanks to data analysis and artificial intelligence, are now able to offer tailor-made itineraries, restaurant recommendations based on food tastes, and local activities that match travelers’ interests. This ability to respond individually to travelers’ expectations not only enhances their experience, but also boosts loyalty.

This summer, travelers expect tourism companies to take full advantage of artificial intelligence to deliver a seamless, personalized and secure travel experience. 41% of French people say they are ready to chat with a chatbot rather than an agent to get recommendations, organize their trip or suggest contingency plans in case of the unexpected. This proportion rises to 58% among 25-34 year-olds and 62% among 18-24 year-olds, demonstrating the enthusiasm of the new generations for these technologies. The integration of these technologies reduces friction during the booking process, cuts the time needed to complete a transaction and improves the overall user experience, making the booking process more pleasant and efficient. Players in the tourism sector who are able to meet these expectations will benefit from a clear competitive advantage, helping to redefine travel standards.

To find out more, discover the success story of our client Goélia, a vacation residence and club company.

1 – Source: Edreams Odigeo
2 – Source: Expedia platform study conducted with Luth Research
3 – Source: Travel Predictions 2022, booking
4 – Source: national federation of institutional tourism organizations
5 – Source: Econsultancy
6 – Source: Zendesk and Yougov study

The human relationship between patient and caregiver ensures well-being and trust in the healthcare system. With the advent of new technologies in the medical landscape, notably Artificial Intelligence (AI), this relationship is gradually being transformed, giving rise to a duality. On the one hand, there is the hope that the integration of AI will free up time for healthcare professionals, thereby strengthening patient support, and on the other, the fear of dehumanizing care. With the launch of the “Digital Health” scheme in 2021, an initiative promoting the emergence of innovative digital health solutions, and the growing adoption of AI by caregivers (54% in 2024)_⁽¹), France aims to position itself as a world leader in e-health. Against this backdrop, how can we ensure that AI serves the interests of both professionals and patients, while preserving the quality of the patient-caregiver relationship?

Integrating artificial intelligence into the healthcare sector

AI offers numerous opportunities for medical teams and care centers. Despite concerns about the possible dehumanization of care, statistics show significant adoption of AI by healthcare professionals. Among its most widespread uses, it facilitates access to medical information (46%), offers specialized training to healthcare professionals (37%) and contributes to the prescription of appropriate treatments (28%)._⁽²)

Its growing integration into medical practices and hospital systems, thanks to technologies such as conversational agents and health monitoring applications, means that patients benefit from access to comprehensive medical information, as well as faster advice. According to Salesforce, 86% of customers prefer to get a response from a chatbot rather than fill in a form on a website, a preference also observed in the healthcare sector. Chatbots offer a personal and anonymous alternative for interacting with healthcare services.

Automate administrative tasks to optimize patient care.

Chatbots are particularly effective in automating and simplifying administrative formalities, as well as providing access to medical information. Thanks to their ability to handle multiple requests simultaneously, bots reduce waiting times, improving the patient experience and boosting satisfaction rates. Accessible 24/7, they offer an immediate, automated response to recurring requests, such as booking appointments, opening hours, issuing medical documents or administrative procedures. It’s also a way of redirecting patients to the right specialty, providing preventive care, and offering initial advice on a wide range of subjects, such as smoking cessation, stress management or healthy eating.

These benefits are not limited to patients, but also benefit healthcare professionals. By considerably reducing the time spent on administrative tasks, and freeing them from these repetitive responsibilities, chatbots enable caregivers to devote more time to listening to and accompanying patients, thus reinforcing the human relationship at the heart of care. In fact, 71% of them would like AI to help them manage administrative tasks⁽³⁾.

However, adopting this technology is not without its challenges and risks. Chatbots are not designed to make diagnoses, nor are they intended to replace healthcare professionals. On a technical level, healthcare chatbots need to be rigorously designed to ensure the accuracy and reliability of the information provided. In addition, there are concerns about data security and the confidentiality of patient information, requiring special attention to ensure transparency and user trust.

The need to protect medical confidentiality

Ensuring the protection of medical secrecy and its ethical use is essential if AI is to benefit both professionals and patients. At present, the deterioration of the doctor-patient relationship is one of the main concerns of 49% of doctors. Familiarization with AI is necessary for its effective use, while making them aware of the associated ethical, legal and social issues. Furthermore, it is vital to ensure that these AIs are used responsibly, by establishing clear standards and regulations to protect patient privacy and guarantee the transparency, reliability and confidentiality of the systems used. To this end, HDS (Health Data Hosting) certification has been introduced to reinforce the protection of personal health data and build a climate of trust around e-health and patient monitoring.

Rather than fearing the dehumanization of medicine, AI should be seen as essential tools for tackling the major challenges facing the healthcare sector. Medical chatbots are a concrete example of the positive impact of artificial intelligence in the medical field. By relieving professionals of time-consuming and repetitive tasks linked to administrative management, medical chatbots complement traditional services, enabling caregivers to focus on their primary mission: offering personalized, quality care to every patient.

Source : AI in healthcare: ally or threat to healthcare professionals? PulseLife (ex 360 medics / Interaction Healthcare) – January 2024

Contributed by Samir Dilmi, Chief Revenue Officer at Dydu

To meet this recurring demand for information, the automation of the most frequently asked questions is gradually becoming the most appropriate response for human resources departments. Synonymous with more efficient management, enhanced business value and an improved employee experience, the use of conversational robots is emerging as a solution to relieve departmental bottlenecks.

More globally, Artificial Intelligence is revolutionizing the HR field. 60% of business leaders plan to enhance their HR departments with more AI over the next 5 years (1). By the end of the year, 69% of routine managerial tasks should already be fully automated (2). With change already underway, how can we continue to maximize the effectiveness of AI to strengthen support for HR departments and simplify recruitment processes?

Conversational robots: the fully-fledged virtual assistant for HR teams

AI and bots are revolutionizing HR processes by automating repetitive and administrative tasks. Leave, health insurance, catering, training, professional expenses, healthcare, teleworking, retirement, employee savings, company life – they support employees from the moment they join the company until they leave. Accessible at all times, on all devices (computers, mobiles or tablets) and via numerous channels (websites, applications, messaging, etc.), they facilitate the employee experience. However, these technologies are not limited to simple optimization. They play an important role in internal communication by directing employees to the appropriate resources, through the creation of virtual collaborators via Teams, for example. Chatbots ensure the effective dissemination of key messages within organizations, particularly those with multiple sites, several business families, or full-remote companies (3). It helps maintain overall harmony in corporate communications, standardizing the company’s discourse and ensuring consistency in the information transmitted to all staff.

Interconnected with the HRIS and the company’s intranet, the bot gives employees easy access to their personal data, to information linked to the convention and framework agreements, and to forms (badge request, canteen registration, etc.). Together, they become facilitators in the promotion of corporate culture.

Recruitment 3.0: the impact of Large Language Models in the talent acquisition process

At least 79% of companies are already using some form of AI or automation in their recruitment and hiring process (4) to make it more automated and interactive. By rapidly analyzing vast volumes of data and CVs to identify the best-qualified candidates, it optimizes and accelerates pre-selection. It predicts candidates’ compatibility with positions, conducts virtual interviews and assesses relevant skills, through the use of algorithms. Employers who have automated parts of their talent acquisition and management processes are very satisfied with the results. Some 2 out of 3 HR professionals report improved recruitment times for vacant positions (5) and resource savings thanks to these technological tools.

Conversational robots offer candidates a more fluid, confidential, accessible and personalized experience throughout the recruitment process. This improves communication with potential employers and understanding of company expectations. In 2023, 58% of candidates said they were happy to interact with recruitment bots during the early stages of the application process (6). They are able to provide information on vacancies, processes and company policies at any time of day, thanks to 24/7 availability. They answer the most frequently asked questions, and personalize advice according to experience and skills, enhancing the overall experience and building trust. Some are even designed to help candidates prepare for interviews by providing sample questions.

The human resources and recruitment market is constantly evolving, requiring employees to adapt. The rapid development of AI and conversational agents in this field reflects a profound transformation in the way companies manage human relations. They bring intelligent automation, increased personalization and operational efficiency, freeing up time for strategic activities and enhancing the overall employee experience. While current figures and projections paint a promising picture, it’s essential to remain aware of the ethical challenges and ensure harmonious collaboration between artificial intelligence, chatbots and professionals to guarantee a balanced and innovative future of work.

If you’d like to find out more, check out our article on the various steps involved in setting up an HR chatbot!

Contributed by Samir Dilmi, Chief Revenue Officer at Dydu

To meet this recurring demand for information, the automation of the most frequently asked questions is gradually becoming the most appropriate response for human resources departments. Synonymous with more efficient management, enhanced business value and improved employee experience, the introduction of conversational robots is emerging as a solution to relieve departmental bottlenecks.

More generally, Artificial Intelligence is revolutionizing the HR field. 60% of business leaders plan to enhance their HR departments with more AI over the next 5 years (1). By the end of the year, 69% of routine managerial tasks should already be fully automated (2). With change already underway, how can we continue to maximize the effectiveness of AI to strengthen HR department support and simplify recruitment processes?

Conversational bots: the fully-fledged virtual assistant for HR teams

AI and bots are revolutionizing HR processes by automating repetitive and administrative tasks. Leave, health insurance, catering, training, professional expenses, health, teleworking, retirement, employee savings, company life – they support employees from the moment they join the company until they leave. Accessible at all times, on all devices (computers, mobiles or tablets) and via numerous channels (websites, applications, messaging, etc.), they facilitate the employee experience. However, these technologies are not limited to simple optimization. They play an important role in internal communication by directing employees to the appropriate resources, through the creation of virtual collaborators via Teams, for example. Chatbots ensure the effective dissemination of key messages within organizations, particularly those with multiple sites, several business families, or full-remote companies (3). It helps maintain overall harmony in corporate communications, standardizing the company’s discourse and ensuring consistency in the information transmitted to all staff.

Interconnected with the company’s HRIS and intranet, the bot gives employees easy access to their personal data, information on collective bargaining agreements and framework agreements, as well as access to forms (badge request, canteen registration, etc.). Together, they become facilitators in the promotion of corporate culture.

Recruitment 3.0: the impact of Wide Language Models in the talent acquisition process

At least 79% of companies already use some form of AI or automation in their recruitment and hiring process (4) to automate it and make it interactive. By rapidly analyzing vast volumes of data and CVs to identify the best-qualified candidates, it optimizes and accelerates pre-selection. It predicts candidates’ compatibility with positions, conducts virtual interviews and assesses relevant skills, through the use of algorithms. Employers who have automated parts of their talent acquisition and management processes are very satisfied with the results. Around 2 out of 3 HR professionals report improved recruitment times for vacant positions (5) and resource savings thanks to these technological tools.

Conversational bots offer candidates a smoother, more confidential, accessible and personalized experience throughout the recruitment process. This improves communication with potential employers and understanding of company expectations. In 2023, 58% of candidates said they were happy to interact with recruitment bots during the early stages of the application process (6). They are able to provide information on vacancies, processes and company policies at any time of day, thanks to 24/7 availability. They answer the most frequently asked questions, and personalize advice according to experience and skills, enhancing the overall experience and building confidence. Some are even designed to help candidates prepare for interviews by providing sample questions.

The human resources and recruitment market is constantly evolving, requiring employees to adapt. The rapid development of AI and conversational agents in this field reflects a profound transformation in the way companies manage human relationships. They bring intelligent automation, increased personalization and operational efficiency, freeing up time for strategic activities and enhancing the overall employee experience. While current figures and projections paint a promising picture, remaining aware of the ethical challenges and ensuring harmonious collaboration between artificial intelligence, chatbots and professionals is essential to ensure a balanced and innovative future of work.

If you’d like to find out more, check out our article on the various steps involved in setting up an HR chatbot!

1 : According to SHRM (Society for Human Resource Management)
2 : According to Gartner
3 : Full remote: telecommuting for all employees
4: According to SHRM (Society for Human Resource Management)
5 : According to SHRM
6: According to Allegis

As generative AI opens up new possibilities in the development of conversational tools, a return to grace for chatbots is yet to come.

Interview given by Samir Dilmi, Chief Revenue Officer at Dydu, for Siècle Digital.

Ah… Chatbots, those services behind which users were promised to converse with some form of intelligence to ask it what they wanted at any time, any day, with a response close to that of a human. What a disappointment this was for millions of Internet users around the world, who quickly lost interest! Will the emergence of ChatGPT and, more broadly, large language models (LLMs) bring brands and consumers back into contact with chatbots?

It’s 2016, and chatbots are invading the internet in leaps and bounds. They can be found on many websites, and especially in Facebook’s Messenger application, which has given them a prominent place. After-sales service, recruitment, human resources, fast food, banking, press, university… In a period when “digital transformation” is hammered into every meeting, chatbots present themselves as a godsend to show that we know how to innovate.

“ Most chatbots have been rather disappointing for end-users, and costly for companies, ” recalls Ghislain de Pierrefeu, Partner Artificial Intelligence and Data at Wavestone. Although some had already put natural language processing into practice, enabling agents to understand multiple intentions, they proved limited in their responses. For those brands who persevered, with the promise of decongesting their customer relations centers, the price was high. They had to mobilize resources to train their chatbots, to provide them with quality and quantity data so that they could correctly read a user’s requests, understand and use specialized vocabulary… ” The players have never managed to process the chain all the way to the end to fetch and integrate personal data. (…)

There was a phenomenon of disaffection. People got fed up and preferred to call an advisor or send an e-mail, ” he adds.

An observation shared by Samir Dilmi, Chief Revenue Officer at Dydu, a French company that pioneered the creation of chatbots. ” This big boom was followed by a big letdown. Many companies jumped on a business challenge without taking care of the product, and without always understanding that it’s not a replacement solution, but a helping solution. “

Faced with a notable lack of interest from the general public, companies have turned to in-house applications, notably in IT support, human resources or sales assistance. “This is where the first HR chatbots began to appear, ” continues Samir Dilmi. In bank branches, advisors were able to access interfaces providing information on all products, and cross-reference them with other data such as account status.

The development of these uses has led to advances in natural language processing (NLP): “today, we’re able to understand a context, a question, and go and find the answer in the right place “. Even if the risk of in-house use is lower, brands have learned to devote time to training their chatbots. However, there are still shortcomings, such as the use of sector-specific vocabulary (banking, energy, healthcare…), or writing at a level of fluency close to human. These shortcomings could eventually be remedied by the integration of LLMs, which would bring additional benefits.

While the first major language models appeared in 2018, it was not until four years later that we discovered the extent of their capabilities, notably through generative AI. For businesses, OpenAI’s GPT, Meta’s LLaMA, or Google’s Bard offer skills that are both ergonomic, because you can ask them anything you like, and you’ll get an answer; and linguistic, because you have to admit they express themselves in near-perfect French.

An obvious advantage of using LLM is the time it takes to train your chatbot. ” The learning curve has been completely overhauled, ” says Samir Dilmi, who draws a parallel with a supervisory method in which a human validates responses, which is still widely used. ” We can retrieve service archives, study conversations, and train the robot before it’s even launched. We save precious time in setting up the bot and absorbing knowledge “.

But be careful not to leave everything in the hands of an LLM, to avoid a well-known but inexplicable phenomenon: hallucinations. These all-purpose AIs can invent answers, and for brands this represents an obvious risk. We should therefore expect to see LLM exploited primarily for simple needs, such as comparing products, creating cooking recipes… Rather than for confirming international insurance cover, or monitoring medical treatment. ” Before, we could be disappointed by a chatbot’s inability to provide an answer, today we could be disappointed by a wrong answer, ” analyzes Ghislain de Pierrefeu.

Castorama lent itself to the game with the launch of a chatbot on its site in November:

Hello Casto. The aim is to ” compensate online for what an in-store expert offers our customers, ” explains Romain Roulleau, Castorama’s Marketing and Digital Director. It operates in a closed environment, in an orchestrator called Athena designed at group level to develop artificial intelligence projects. It deals with issues relating to security, the relevance of responses, and the moderation of exchanges. Hello Casto is based on the three main LLMs on the market, Bard, GPT and Claude (Anthropic), who intervene according to the question and context. Despite encouraging feedback, the company is cautious. ” We make corrections several times a day. (.) We know that hallucinations do exist, and that’s why Athéna is here, ” adds Romain Roulleau.

As chatbots mature and technological solutions become available, a return to favor is definitely on the cards.

Artificial Intelligence, popularized by ChatGPT, is playing a growing role in many fields, and the market is expected to reach $1,300 billion by 2032. AI uses models such as GPT-4 to generate written, artistic, and musical content, as well as computer code.

In 2023, 72% of Internet users used generative AI to search for a product. It has also become a valuable resource for content creation and is revolutionizing how companies interact with their customers, manage operations, and create value. But how can we control the content produced and ensure it is accurate? Or measure its environmental impact? What major challenges can we expect in the coming months?

Generative AI: an essential partner in the modern world of work 

Virtual assistants and AI have revolutionized our ways of working and become essential partners in boosting efficiency and productivity. They simplify tasks and produce rich, varied content (images, videos, texts, translations). Companies are adopting conversational agents, or chatbots, on a massive scale to provide quick and accurate answers, reduce wait times, and increase customer satisfaction.

However, it’s important to remember that these algorithms can only create from what already exists. Employee expertise is essential to enrich information databases and create emotions. AI should be able to adapt to changing consumer preferences in real-time. Companies must, therefore, implement systems that learn and evolve with customer needs, which requires greater agility and responsiveness.

The real potential of these new technologies lies in human-machine collaboration. This synergy will be key in future years. However, it will only open the door to new professional opportunities with continuous employee training and constant adaptation.

New 2024 challenges to harness the power of this technological revolution

The rise of AI raises several financial, environmental, regulatory, and security issues. New artificial intelligence systems require a responsible framework to regulate the sector and prevent misuse.

The security of AI systems is a major concern. Chatbots can be vulnerable to adversarial attacks, and model errors can lead to the dissemination of false information. In 2023, 4 out of 10 consumers cited data privacy as the main reason preventing them from using AI for online shopping advice, followed by concerns about the reliability and quality of the answers. The only way to overcome this challenge and guarantee accurate information is with cybersecurity research and robust models.
There are also financial challenges. Large Language Models generate significant costs, requiring the creation of new business models. Investment in research and development and the acquisition of advanced technology is imperative. Companies must consider the cost of data storage and management, customized products and services, maintenance, updates, ethics, and compliance to develop effective, sustainable strategies in this new, constantly evolving economic landscape.

Greenhouse gas emissions (server powering), creation and management of electronic waste, water consumption, hardware replacement… Companies looking to use these systems face many challenges. AI models, in particular deep neural networks, require massive computing capacity. The servers and data centers powering these systems consume a great deal of energy, significantly contributing to a company’s carbon footprint. In an article that should be published later this year, Professor Ren’s team estimates that ChatGPT ‘drinks’ 500 milliliters of water, the equivalent of a small 50cl bottle of water, for every 5 to 50 questions asked.

In 2024, the challenge will be a more in-depth and sustainable integration of these technologies into businesses’ day-to-day operations. This will boost efficiency while limiting misuse. Globally, only 12% of companies are capitalizing on AI to improve their performance, and they are growing up to 50% faster than competitors. Adapting to these new technologies and overcoming the subsequent issues will enable companies to offer tailored experiences, as well as optimize their operations, and guarantee data protection. Awareness, innovation, regulation, and data measurement and analysis will be key to overcoming these challenges.

In the constantly evolving digital age, customer relations are being redefined through the dynamic integration of chatbots. These intelligent conversational agents, powered by artificial intelligence, are transforming customer relationship management. Join us as we explore the significant role chatbots are playing in this changing landscape, as well as the benefits, evolutions, and prospects of this digital revolution.

Using Chatbots to Boost Customer Relations

The Evolving Customer Relationship in the Age of Digital Transformation

Digital transformation has reshaped the way companies interact with their customers, opening up a wide range of possibilities. Customer relations chatbots  have become key assets to support this change. From simple information searches to order tracking, claim management, or after-sales issues, chatbots provide instant 24/7 support on websites and mobile apps. Their constant availability is perfectly in line with the age of instant gratification, meeting customers’ expectations for immediate and personalized answers about products and services. 

Companies can identify the most frequently asked questions and delegate them to their chatbot, enabling customers and prospects to get quick answers that meet their needs. Chatbots can also be interconnected with internal software such as CRM, to create a fully personalized experience. What’s more, integration with livechat technology enables to hand-off to a human operator if necessary, while retaining the dialog history. Phone (callbots) or voice (voicebots) versions enhance accessibility, ensuring an optimal customer experience. 

The Central Role of Chatbots in Customer Satisfaction

Chatbots play a key role in meeting new customer expectations. An estimated 67%* of consumers in the world have interacted with a chatbot over the last 12 months. Chatbots provide instant answers by using customers’ favorite communication channels, such as instant messaging systems. Artificial intelligence also enables chatbots to understand customers’ natural language, analyze their needs, and provide personalized responses. 

This kind of customization has a significant impact. It makes customers feel understood and cared for as individuals, boosting satisfaction. Chatbots are precious allies in the quest for more efficient customer relationship management. They reduce customer waiting times, improve the quality of responses, and optimize the handling of requests. Adopting a chatbot helps build a smoother customer journey, improves satisfaction, and fosters loyalty. Chatbots are particularly well suited to sectors such as e-commerce or banking and insurance, among others. These tools have become a must-have for companies seeking to optimize their customer relationship management.

The Benefits of Chatbots for Businesses and Customers

Benefits for Companies

Chatbots offer customers a wide range of benefits that enhance their experience. But what exactly are these benefits?

• Increased responsiveness: chatbots are available 24/7 and respond instantly to customer questions, eliminating waiting times.
• Personalization: API connections and CRM tools enable chatbots to provide tailored answers, integrating specific data. This combination creates more personalized interactions, for a one-to-one customer experience.
• Easy to use: chatbots use natural language which simplifies interactions.

Benefits for Companies

Chatbots also offer companies significant benefits:

• Task automation: chatbots automate repetitive tasks, freeing up employee time.
• Cost reduction: by automating frequent requests, companies can reduce operating costs
• Improved productivity: chatbots help human agents by providing contextual information, making them more efficient.

A Combined Future for Chatbots and Customer Relations?

The Evolution of Chatbots Towards More Human and Personalized Interactions

Chatbots are constantly evolving to offer ever more human and personalized interactions. The integration of artificial intelligence into these systems enables a deeper understanding of natural language, providing more authentic exchanges with customers. Some chatbots can even detect emotional nuances in conversations and adapt their responses to the customer’s emotional state. The aim is to create interactions where customers feel genuinely cared for in an empathetic way, thus strengthening customer relationship management.

Integrating Chatbots with Other Communication Channels and Advances in Artificial Intelligence 

The future of chatbots lies in their ability to integrate seamlessly with other communication channels. For example, a customer can start a conversation with a chatbot on a website, then continue their discussion on a messaging app. Advances in natural language understanding, personalized recommendations, and the anticipation of customer needs enable chatbots to manage the customer relationship more effectively. This combination of seamless integration and continuous improvement in artificial intelligence makes chatbots a key part of the customer relationship. 

To find out more about how chatbots and artificial intelligence are revolutionizing the customer relationship, download Kiamo’s complete white paper, where you’ll also find an interview with Samir Dilmi, Chief Revenue Officer at Dydu, as he shares key insights on this topic. 

¹Chatbot recruitment: the benefits of conversing with applicants 

Since the arrival of ChatGPT by OpenAI at the end of 2022, advances in AI have brought language models to the fore and accelerated their development. Large Language Models (LLM), through the use of Deep Learning, have radically transformed the way companies tackle language-related problems and offer a wide range of benefits. How do these generative artificial intelligence tools help companies modernize and improve their performance ?

Facilitating In-house Processing and Analysis of Textual Data

Companies today are undergoing a digital transformation and are looking for the best in terms of efficiency and performance. Businesses receive a constant flow of unstructured data from multiple sources such as emails, reports, customer reviews, or social media. Transforming all of these documents into knowledge bases is a particularly time-consuming task. One of the main strengths of Large Language Models lies in their ability to process, understand, and analyze large volumes of textual data. Large Language Models convert this data into actionable information, facilitating informed decision-making.

Automating Repetitive Tasks to Save Time and Energy

Large Language Models can be used to automate many repetitive language-related tasks, such as writing reports, classifying documents, responding to customer queries, and even translating text into foreign languages. By relieving employees of these time-consuming tasks, companies can optimize efficiency and focus resources on higher added-value activities.

Enhanced Customer Experience Through Better Understanding of the Human Language ?

Generative artificial intelligence processes huge volumes of data, providing more efficient analysis and task precision. The algorithm understands, synthesizes, and classifies information based on several parameters in order to provide the most accurate answers. By understanding the nuances of human language, it can analyze customer requests, needs, and feedback in detail, carefully identifying the services required and offering better customer service.

LLMs also enable companies to create or enhance chatbots and intelligent virtual assistants capable of delivering precise and personalized responses to customers. These systems provide 24/7 customer service which improves the customer experience and boosts overall consumer satisfaction.

Promoting Innovation and Creativity in the Workplace

Large Language Models are also powerful tools for encouraging innovation within companies. They can be used to drive new business approaches and even help with product design. They can spark inspiration and help employees be even more creative.

Fostering Strategic Decision-Making

In an ever-changing world, companies are faced with complex strategic decisions. Large Language Models provide in-depth analysis and predictive insights based on the analysis of large text datasets. This enables executives to make more informed decisions and better anticipate market trends.

Large Language Models are a strategic asset for businesses, offering them a better understanding of textual data, efficient automation, an improved customer experience, opportunities for innovation, and more informed decision-making. Companies that embrace these emerging technologies will be better equipped to succeed in a competitive and ever-changing business environment.

Feel free to reach out if you want to find out more about the integration of generative AI into our solution !

Over the last ten years, conversational agents (also known as bots) have gradually become a key part of the customer experience.

Automating customer interactions presents several benefits for companies: time savings, efficiency, cost optimization, etc.

Although self-care tools are particularly popular with consumers, the image of robots vs. humans persists in some people’s minds. 

At a time when brands need to connect with customers to build loyalty, how can they implement a winning strategy that leverages technology while maintaining a human touch?

Automated tools: the answer to changing consumer behaviour

Digital tools have brought about big changes in how customers consume and relate to brands.

The rise in multichannel platforms has led to a surge in incoming requests.

Consumer habits have also shifted, with people now shopping at any time of day and night, on public holidays, etc. 

As customer service budgets get smaller, companies must provide users with a 24/7 service that can assist them and provide immediate answers.

Customer service professionals admit that responding to all requests adequately is difficult. In fact, 63%* say that they struggle to strike a balance between speed and quality¹. 

72%* of consumers also prefer to solve problems themselves rather than speak directly to a salesperson or customer service rep².

With this in mind, automating the customer journey seems the best solution for satisfying users and customer service.

Technology and people: the ideal combo to ensure optimized customer service?

Contrary to popular belief, companies don’t implement bots to replace people but to add value for end-users and lighten the load for support teams.

Automating the processing of recurring requests, or pre-filtering them, saves advisors precious time, which they can use to reduce waiting times and personally handle complex requests.

Employees can then enjoy helping customers again while customers get the answers they need, without having to wait.

Bots, a bridge for businesses?

Conversational agents are designed to complement customer relations and take over from operators. They are available on various platforms, including social media, instant messaging or e-commerce sites.

Gone are the days of the little window at the bottom of your computer screen. Today, consumers interact with chatbots via Messenger or WhatsApp.

Combining artificial intelligence and human intervention enables customer services to provide a personalized and seamless experience.

This alliance boosts customer satisfaction and helps to handle incoming requests without overloading the support teams.

Let’s take an example. A car insurance bot can provide clear and quick answers to questions about policies or deals. 

But if a policyholder needs immediate assistance after an accident, the bot can connect them to an operator.

For a garage, an automated solution provides a contact point outside opening hours.

Customers can, therefore, make an appointment to have their car serviced or their windshield repaired at any time of day or night. In this case, the bot serves as a bridge for business.

The challenge is to strike the right balance between the automation of certain tasks and human support.

A solution equipped with natural language processing (NLP) will provide customers with a seamless experience by quickly understanding their requests, no matter the phrasing.

Finally, we highly recommend regularly collecting reviews and customer satisfaction scores to adapt to their expectations.

Natural language processing and a solution that can hand requests over to an operator when they become too complex are therefore key to automating the customer relationship without dehumanizing it.

Providing relevant answers in the shortest possible time will make customers feel valued and important.

¹ Salesforce: Focus on customer service 2020 https://www.salesforce.com/content/dam/web/fr_fr/www/PDF/guides/state-of-service-fr-2021.pdf

² According to a Forrester study

Since the Covid-19 health crisis, we’ve entered a new era of consumption, where more and more purchases are made online.
Consumer habits are changing, and customers are becoming increasingly demanding.
They want to consume better, everywhere, all the time.
Brands must be more mindful of the customer journey to increase satisfaction and their conversion rate.

Find out how a chatbot can improve your users’ experience.

E-Commerce Chatbots as Customer Relations Allies

A Seamless Customer Journey

• An e-commerce chatbot is available 24/7 to assist users throughout their buying journey
• It provides automatic, quick, relevant, and personalized answers
• It removes user barriers and buying blocks 
• It boosts sales 

Less Pressure on Customer Service Teams

Chatbots are a customer service agent’s best friend because they:

• Manage recurring questions and tier 1 problems
• Free up time so that agents can focus on more complex issues
• Reduce customer service waiting times

To find out more about the benefits of our e-commerce chatbot solution, read this article by our Head of Marketing and Communications, published by JDN.

Workplace inclusion is at the heart of CSR policies and increasingly held to high standards. Inclusion should be a key part of any business’ global strategy in order to be better valued longer term by employees and customers alike. People often think of bots’ skills as being diametrically opposed to human qualities. However in this particular case, implementing a bot can actually help companies move towards a more egalitarian and unprejudiced workplace.

Developing a neutral approach to requests

The notions of equal opportunities and accessibility are regularly called into question at work. A study* revealed that 57% of French people strongly believe that equality in the workplace does not exist, which comes as an alarming assessment. To overcome this challenge, some companies have chosen to include conversational bots in their strategy to act as neutral agents. These selfcare tools – callbots, chatbots, and voicebots – handle requests from employees, customers or applicants in the most objective way possible. They are instantly available, regardless of the user’s request, vocabulary, appearance or background. Bots are never judgmental. They are programmed to understand natural language (including colloquialisms, SMS, or spelling and grammar mistakes) and answer each request without bias or value judgement. The fact that they handle each request equally does not compromise the quality of their service. For example, recruitment bots enable companies to manage CVs in a completely impartial manner, based only on the skills required for the job.

Promoting digital accessibility for everyone

Bots are visible, easy to use and compliant with French accessibility RGAA guidelines. They can even go one step further with a voice feature, which is particularly inclusive for people with disabilities or anyone who is not at ease with digital tools.

Bots are available 24/7 and provide customers and employees with direct, consistent and standardised support. If the bot doesn’t understand a request, they can suggest an alternative or rephrase the question and, if necessary, transfer the user to a live agent. Bots handle the bulk of incoming requests, so physical agents can spend more time on cases that require specific treatment and provide personalised and more inclusive support.

Making users more autonomous

In addition to handling information in an inclusive way, bots also encourage users to be autonomous. For example, a recruitment bot can help applicants manage and monitor part of the process themselves, enabling recruiters to focus on processing their application.

Another example is after-sales bots. They can use fault codes to help customers solve their problems quickly without an operator’s intervention. These tools make an employee’s day-to-day less repetitive, which increases satisfaction and well-being at work. In many cases, bots also improve the company’s image in employees and customers’ eyes. According to a study**, 73% of HR professionals believe that implementing a company chatbot is an asset for the employer brand.

Bots help employees focus on higher added-value tasks, which in turn makes them feel more valued. This continually improving and inclusive technology focuses on each individual and treats everyone equally. 

*Study conducted by Qapa (Adecco) for Equal Opportunity Day
**4th edition of the barometer of the use of chatbots in HR

People often underestimate robots, yet they play an increasingly large role in our lives, especially at work. For example, more and more customer service departments use callbots to improve their performance.

Orders, refunds, bookings, insurance… Lots of queries require an immediate customer service response. However, there aren’t always readily available agents, which can be uncomfortable for customers as well as employees, who have to manage dissatisfied calls and messages on a daily basis. This situation can soon have a negative impact on the company. Could the solution be… a robot? Callbots are available immediately, all day and night, and can prove to be a key customer service asset.

A Robot that Adapts to Customer Needs

Call centres are rarely open all day, every day. Which is why some companies provide a complementary service in the form of a callbot. In a study* on the most important things in customer service, 60% of those surveyed highlighted quick problem solving, while 40% valued 24/7 assistance.

Callbots handle 100% of calls, significantly reducing the number of answered queries and unavailable agents. The tool therefore enables to capture all leads and respond to all existing customers. Everyone benefits: companies can seize new business opportunities, and customers don’t have to adapt to specific opening hours.

More Efficient and Happy Customer Service Reps

Customers and employees alike appreciate callbots, especially customer service agents who receive a lot of incoming calls. The robot handles customer requests first and then transfers those requiring expertise to an employee. Reps therefore only deal with part of the call and don’t get as many. The callbot authenticates the caller, verifies the order number etc., before transferring the call, thus ensuring quality interactions and better request management.

This process makes the employee’s work more rewarding and adds value, which helps develop their skills. The customers they deal with are also happier because the bot directs them to the right person and has already identified them. Callbots simplify employees’ day-to-day work and improve its quality, which makes the job more appealing. This in turn helps reduce staff turnover, which is often very high in customer service.

A Positive Financial Impact on Companies

A callbot creates value for customers in a modern and adaptable environment. The benefits of implementing a robot can soon be seen through the company’s data. Their recruitment needs will look different since the bot can handle certain calls by itself, requiring less, but more qualified, employees. They can also calculate the callbot’s ROI, almost in real time, based on the number, duration and resolution rate of calls received, the number of calls handled fully/partially, etc. 

Finally, this tool can provide businesses with a competitive edge over those without one. Depending on the context, callbots can manage 100% of customer requests outside working hours. This enables companies to place customer needs at the heart of their strategy and improve satisfaction, while building loyalty. 

* Source: Vocal Com

We live in a highly digitalised world, in which modern technologies have become a way of life, and the collaboration between humans and robots a must. This is increasingly true in the workplace. Behind every user/consumer, there is often an employee who would like to benefit from the same technology at work. Employees, managers, HR heads… Everyone has something to gain from implementing robots in the workplace, particularly a HR chatbot. 

Combining Immediacy and Personalisation

Implementing an internal company chatbot improves employees’ quality of work life. It provides instant access to information 24/7 wherever they are and from any device (computer or mobile, website, app, desktop, company messaging system, etc.) on a wide range of topics: HR, IT, purchasing or even general services. 

A chatbot treats everyone equally, whatever their status or request. It understands natural language and helps tackle inclusion issues that are an increasing focus for companies. A chatbot can be customised with tailored answers through the use of standard variables or by connecting it to the company’s HRIS.

The tool therefore frees up time for employees/managers and HR to listen and talk to each other. This is all the more important during health or economic crises. According to a study, 70% of HR professionals believe that chatbots are even more useful in times of crisis. By performing time-consuming tasks, the chatbot frees up time for each department to focus and think about the new and engaging issues that arise during such periods.  

Improving Employee Wellbeing and Productivity

A chatbot can play a key role in a company’s global strategy. It is a great support for businesses with a wide range of profiles, remote sites, a high turnover and/or travelling employees. PwC (specialised in consulting, auditing and legal and fiscal expertise), for example, has a dense intranet and wanted to facilitate the onboarding of new employees (nearly 2,000/year). They therefore implemented an internal chatbot to create a connection and ensure consistency. We often come across this need in decentralised companies, who use chatbots as a unifying internal communication tool.

The benefits of implementing this technology, which is more and more widespread, are immediate. A good level of service has a positive impact on employee wellbeing as well as productivity.

Adapting to the Challenges of Workforce Transformation

Some employees may be wary of the implementation of selfcare tools, particularly robots, for fear that they will dehumanise interactions in the workplace.

Yet bots are increasingly “close” to their users. For example, they can chat with employees on internal messaging systems (Teams) like any other user. When the human touch is required, the chatbot redirects users to the right department or person, meaning that everyone has more time and energy to build a good relationship.

Contrary to popular belief, not all robots take people’s jobs. In fact, the opposite can be said of chatbots. The disappearance of certain tasks allows new ones to emerge. Indeed, while 59% of HR professionals believe that chatbots will replace certain functions, 69% say that they will create new jobs.

* Source: Barometer of HR chatbot use by dydu 

* Source: Client story – dydu

Bots can be valuable tools to lighten the workload and enable employees to be more autonomous and efficient. They are easy solutions to implement and facilitate corporate life and management.

Bots Are Employee Allies

When a department’s workload becomes too heavy, the first question to ask is, what can be automated? Which recurring tasks are particularly time-consuming? Leave requests, questions about remote working, multiple reminders about an internal survey, questions about using a new tool, etc. These are just some examples of tasks that conversational robots (chatbots, callbots, voicebots) can handle.

Another important point is anonymity. As an employee, you may not dare ask your manager certain questions. Whereas bots can answer instantly, in the most neutral way possible, and redirect you to the right department if necessary.

Bots can also be used as an assessment tool, for example to measure the percentage of questions on one topic rather than another. You can then adapt your management accordingly and highlight certain subjects via other communication channels. If you get a lot of questions about remote working, for example, it may be worth drawing up guidelines, in which you explain the terms and conditions, and communicating widely on the subject internally.

An Adaptable Tool that Supports Change

The new working methods that have emerged from the health crisis – full or partial remote working – have highlighted the usefulness of bots. Tools that are available 24/7 can also be particularly helpful for organisations with employees working off-site, in different countries, or on shifts. Or for companies that regularly call on contractors and want to grant them access to certain bot contents.

Chatbots are flexible tools, and their content can easily be adapted to any type of change, from a merger with another company, to an environmental disaster, or a social conflict at work. Whatever the situation, bots enable you to adapt and continue to provide quick answers, limiting big and rushed changes, like during the health crisis. As such, in the midst of a lockdown, 70% of respondents who use chatbots reported that the tool was even more useful in times of crisis (*).

Although bots can adapt easily and cover a wide range of topics, they cannot replace employees. If you ask HR professionals if they think this technology will replace certain jobs, 59% answer “yes”. But, if you rephrase the question and ask if they think that bots will create new jobs, 69% also answer “yes”*. Robots and employees can work together. By automating recurring questions, chatbots make employees’ day-to-day work lives less repetitive. As callbots, chatbots and voicebots become more sophisticated, time-consuming tasks are increasingly reduced, in favour of more complex and challenging actions that boost employee efficiency. 

*According to a dydu Chatbot Observatory. 

Find out more about our employee chatbots.

When it comes to technologies using artificial intelligence, time is an essential aspect of their algorithm performance. Indeed, the concept  behind these solutions is that they are self-learning, requiring more or less assistance. Getting a conversational robot (chatbot) to perform and train quickly involves several elements.

The Chatbot’s Foundation: NLP Performance 

The natural language processing engine, or NLP, is based on, and integrates, several elements that are more or less complex. The most basic engines only focus on one word. This keyword recognition system is very similar to a search engine.

At dydu, we’ve been developing and optimising our algorithm for over 10 years. When a user asks a question, our engine analyses all the words in the sentence. It corrects spelling mistakes and identifies the root form of a word, as well as its synonyms. This enables the chatbot to understand lots of different ways of expressing the same intent (or question).

The Chatbot’s Framework: Supervised Learning

Most clients create a traditional chatbot from scratch, building a knowledge base according to their business and use cases. A client setting up a chatbot to handle rail transportation issues won’t have the same needs as one looking to provide IT support for employees. Knowledge bases, i.e. all the questions and answers, are specific to different audiences and scenarios, as is the vocabulary. Once the questions and answers have been defined, time needs to be spent on configuring and training the chatbot. This involves thinking about different ways to formulate the same question. And also testing and finetuning the bot repeatedly to optimise its understanding.

Given that conversational robots speak on behalf of brands to potential and existing customers or employees, we decided, at dydu, to opt for a supervised approach. Either way, humans have the last word and decide, not the robot.

Specific tools in the chatbot’s admin console make this calibration easier and quicker. The engine continually suggests matching optimisations and improvements based on its knowledge base:

• Firstly, by sending alerts in the event of close knowledge articles.
• Then by highlighting misunderstood sentences and, most importantly, suggesting alternatives (to be confirmed or rejected). The bot manager can add the reworded knowledge articles or complete the matching group in a click.
• And finally, by performing quality alerts and reporting any answers with errors, such as broken links.

These preventive and auditing features are therefore key to your bot’s ability to be trained effectively.

As far as the process is concerned, we recommend using samples of real testers from the start with different profiles.

Ready-to-Use: Pre-Trained Knowledge Bases

With our Expert chatbots, all of this calibration work, which can take time at the beginning, has already been completed.

We’ve already defined the themes, questions and matching answers. We’ve also grouped the sentences that mean the same thing together. For example, this is what enables our Council chatbot to understand the following terms equally well: canteen, school lunches, cafeteria. 

For our HR chatbot, dydu has created and tested 110 matching groups associated to 300 knowledge articles.

Creating a bot from an existing model therefore allows you to benefit from all the work that has already been done. This will help your teams quickly adopt the content, by only focusing on the customisation elements specific to your company. Your users will immediately benefit from an excellent experience.  And it will be even easier for you to continue optimising your chatbot in the future!

Standard Reports: a Key Part of the Solution

When using the Dydu solution, you can instantly view your bot’s key performance indicators at any time and over any period. Our BMS (Bot Management System) includes an entire section for this exact purpose, called Analytics.

It includes information about the number of visitors and the number of dialogues. More specifically, you can see the number of interactions per dialogue and also the quality of these interactions and dialogues. This is determined based on the bot’s understanding of the questions and its ability to provide a satisfactory answer. You can also immediately and visually identify your users’ areas of concern by looking at the most frequent topics discussed with the bot.

In order to continually meet expectations, we conducted an exploratory study in Autumn 2020 with 10% of our clients, to fully understand their needs in terms of analysis, as well any gaps between their current use of reports and a potentially more advanced use of their data. We turned the expression of these needs into specifications, enabling our developers to provide a solution that is now available as standard in our software.

Before going into further detail, let’s take a look at who uses bot analytics and why.

Use of Analytics: Who Does What?

Content managers focus on the knowledge base (questions, answers, matching groups, modes of expression within the interactions and dialogues, etc.). They check the analytics and conversations to correct and enrich the bot, so as to answer even more questions correctly. Their objective is to ensure that the bot provides users with as much satisfaction as possible.

Project managers manage the bot’s different stakeholders within a governance framework. They are responsible for justifying the bot’s budget and measuring its ROI. They also work with the business or department to optimise internal processes and make optimisation recommendations. Finally, they often use analytics to manage, or even pay, one or more service providers. The project manager therefore uses analytics to achieve objectives – validation, areas for improvement -, and also communication (evangelism and building loyalty). Bots are also selfcare tools, so project managers often use analytics in a more global context: either by comparing the bot’s performance with other selfcare tools, or by consolidating them into a more complete user experience. 

According to our analysis, the 3 main analytics use cases are, on an equal footing: monitoring bot use, identifying and prioritising actions to carry out and internal communication.

For More In-Depth Analysis

The majority of our clients use existing dashboards, but some work with data outside the BMS, to carry out more in-depth analysis. For example, they use the dialogue content to identify their users’ characteristics (leads, customers, etc.) and expectations. Using the details of the interactions, they can optimise their journey and improve their conversion and satisfaction rates.
In addition to standard reports, some of our clients were therefore having to reprocess data as a result of scattered or missing KPIs.

To facilitate this work, Dydu completely reviewed the available data exports. They are now organised into 2 clearly identified areas: dialogues and interactions. The available fields in each of these areas have been significantly enriched. As such, there are now 22 fields for the dialogue export (vs. 6 previously). For example, we added critical elements to understand the bot’s use: contextual elements (channel, browser, etc.), as well as dialogue-specific elements, such as start and end time, length, number of interactions – overall, business, social -, requested and provided satisfaction. We have also add the type of dialogue, which allows to distinguish between production dialogues and test dialogues.

Moreover, we’ve added the position of the interaction to the interactions export. This data is particularly useful for analysing a churn funnel, for example, and to provide better answers earlier on. We have also inserted a Bot UUID field, which is essential for Metabots, to analyse each child bot’s interactions. We have also added a hierarchy (knowledge path in the base), which is particularly useful in rich and complex knowledge bases, to identify the most frequent and satisfactory (or not) themes or sub-themes.
These can be found in the BMS, in the Dialogue menu.
The Dialogue and Interaction exports are available now. We’re also working on a Knowledge export which will be available soon.

For More In-Depth Communication

The main benefit of these Excel exports is that they can be processed in any Business Intelligence tool and can therefore be used for all types of analysis and combined with data from different apps or sources.

Since these tools often have advanced data visualisation capabilities, it is easier to create powerful visuals in the company’s graphic identity. Project managers can thus communicate on bot-related performance and actions more easily, and with greater impact. 

Remember that exported data depends entirely on your bot’s data retention settings. If you have chosen to anonymise data after a certain period, not to retrieve the user’s IP address, configure the GDPR in a certain way… then you will only retrieve the data from the dialogues that you’ve chosen to keep. It is worth taking this into account when defining the frequency and dates of your exports.

The technology behind our solution’s NLP (natural learning processing) engine and conversational strategy are key to providing a true conversational experience. The AI, which provides you with recommendations to monitor and optimise your bot’s content, also enables you to constantly adapt to changes in your customer relationship. For the most part, you are fully in charge of your content, which depends on your industry, job and the specific use cases you want to cover.

Some words can have several meanings. For example, “ticket” can mean something completely different depending on whether you work in the finance, transport or culture sector.

However, drawing on our experience, we have successfully developed content modules that enable us to implement bots very quickly, for two reasons:

• The content is shared by a large number of organisations. Whether in terms of potential questions and associated answers

• This content has already been tested and optimised

Furthermore, for highly regulated industries, the same framework can be applied to most answers, so there is no need for companies to search for this shared information. 

These expert, “ready-to-use” chatbots are particularly useful for sector-specific or vertical use cases.

Let’s take a look at two examples.

Council Chatbots, for a 24/7 Relationship With Residents

Local residents’ concerns and preoccupations tend to be similar from one place to another, whatever the size of the town or village. However, councils don’t all have the same means to respond to constituents. An expert council chatbot is ideal for providing easy and permanent access to everyone at any time of day.

Which led us to create a bot based on 17 recurring themes :

• Administrative procedures: change of address, marriage, electoral census, etc. This is the broadest theme.
• Waste management
• Mobility, including parking and transport issues
• Community life
• Social assistance
• Employment
• Children and education: schooling and after-school activities
• Crisis management
• Disability
• Housing
• Daily life
• Leisure, sports and culture
• Mediation and justice
• Security
• Senior citizens
• Town planning: general and specific requests, administrative procedures
• Council contacts and useful information

Overall, more than 230 questions and answers, phrases and reworded phrases are available for use by 35,000 French councils. This technology aims to make local residents’ day-to-day life easier.

By way of comparison, 80% of a customer relations bot‘s traffic calls on an average of fifty knowledge articles (*).

60% of this content is “ready to be published”. For the rest, the answers are tagged “to be confirmed”, to be completed or personalised.

HR Chatbots, for a Seamless Employee Journey

An expert HR chatbot can provide employees with immediate and consistent answers to recurring questions from day 1, wherever they are and whatever the time.

Once again, we have predefined more than 250 knowledge articles (question – answer pairs) on 20 common Human Resources themes: company information, benefits, leave and absences, employment contracts, yearly appraisals, family and health, training, management of arrivals and departures, internal mobility, pay, working hours, remote working, general questions, etc.

Employees can access the bot from any work device, on the company’s intranet or collaborative messaging system (e.g., Teams). This relieves pressure on HR, as well as managers, who can then focus on more complex tasks and their core function.

Either way, these tools are particularly inclusive because users can use their own words and expressions. They’re not afraid of asking questions and receive consistent information in return. Implementation is also accelerated because you only need to focus on your company specificities, safe in the knowledge that the rest of the content has already been tried and tested.

(*) Source: internal H1 2021 dydu statistics.

Are you looking to automate simple and repetitive calls with a callbot? While chatbots tend to be more generic, virtual switchboards meet specific use cases, such as signing up to a contract, reporting an incident or tracking an order. Depending on your pre-defined strategy, a callbot can handle end-to-end calls or pre-qualify a request and transfer the customer, and their conversation, to an operator. This article takes a look at the different stages involved in implementing a callbot with dydu.

Callbot Objectives and ROI

The first step before launching a callbot is to precise your objectives, your end-users’ profiles and their needs. This will enable you to define the bot’s scope and use case(s). Callbots, can be used to assist or partially replace call centers. The processes and scripts are therefore often largely pre-existing, which makes it fairly easy to define the bot’s scenario(s) and potential savings.

To ensure that the project is profitable, it is essential to establish certain KPIs from the outset. You can assess your ROI based on hypotheses of time management and savings rates. Let’s take for example an insurance claim scenario. If the callbot is in charge of checking the user ID before the conversation begins, then you can look at the percentage of correct authentications in wich timing… You can also compare and extrapolate with exclusive use of the callbot outside of business hours.

Customer Flow and Callbot Technical Requirements

The next step is to define where the callbot fits into your customer flow and any technical requirements. Ask yourself the following questions:

• Which number will the callbot be reached on? At what times?
• Will the callbot receive calls directly or will there be an intermediary IVR (Interactive Voice Response)?
• Will the caller be redirected to a human operator?
• Should any information from the conversation be passed onto this operator?
• How many simultaneous calls should you expect?

These questions are usually analysed during the commercial phase. The next stages focus on creating and integrating the bot.

User Journey and Data

Before building your bot’s knowledge base, it is important to define your user journey and the data you want to collect , thru two workshops.

Use Case Workshop

Callbots meet very specific use cases, such as reporting or checking the status of a claim. We therefore recommend creating a flowchart for each use case to establish the customer journey on the phone. This flowchart should include:

• The exact text spoken by the bot : this will enable you to check that the conversation flows smoothly. For example, asking “what day did you make your request?” rather than “request date” allows to ensure that the user knows what to do and what kind of answer to give.
• The answers expected by the bot : such as “yes” or “no”, as well as the structure of certain data required for calls to web services. For voice recognition, this often involves numeric data (phone number, licence plate, customer ID, file number, etc.) or alpha data (family name, type of problem, etc.).
• Decision Trees : this allows to identify what answers to expect from users. For example, if the callbot asks if the user knows their file number, you should provide a “yes” and a “no” branch. And maybe also an “I don’t know” or “where can I find my file number” branch.
• Interconnections with a webservice with data to be transmitted and values to be retrieved : this allows to see if the use case can retrieve all the necessary information and the structure of the expected values. Similarly, you can prepare use case answers with the list of possible responses.
• Escalation to human operator if applicable : in which situations and with which information.

At dydu, we support our clients with a project manager during the entire project. They also prepare the flowchart with you.

Analytics Workshop

Callbot conversations have a defined beginning and end. You can obtain analytics on completed dialogues, conversations where the customer hung up before the end, which use cases were involved, why the calls were made, etc.

This workshop helps to define the variables to add and the data to use as variables as well as the structure of the analytical report to be put in place for an optimum monitoring.

Building the Callbot Knowledge Base

Your use cases, spoken words and variables are now fully defined. You can start building your callbot’s knowledge base!

Integration of Knowledge Articles

At dydu, we’ve decided to let our conversational AI experts integrate the knowledge and not our clients. We configure the decision trees, knowledge articles and spoken phrases in our Bot Management System according to the flowchart approved during the previous phase. In most cases, clients can change their answers autonomously.

We also add sentences and matching sentences (groups of words and expressions that mean the same thing). The aim is to maximise the bot’s understanding.

Voice Design

The next step is to choose the voice of your callbotfrom a catalogue. Depending on the TTS (Text to Speech) provider, the client can also record a voice of their choice.

In some cases, we add SSML (Speech Synthesis Markup Language) tags to the knowledge articles. They allow to specify punctuation and pronunciation in the Text to Speech engine, the speech transcription brick. It is then possible to add pauses, read numbers or letters separately, read dates and times, and provide pronunciation instructions (for example, to read more slowly). It is also possible to manage blanks in the conversation and to automatically prompt the user.

The vocal aspect is important to make conversations as smooth and natural as possible for end-users.

Test & Learn and Release of Callbot

This step involves testing the callbot’s voice on the phone with a limited panel. If possible, the members on this test panel should be quite different (age, gender, accent, etc.) and use a wide range of wording to test the bot’s understanding.

We provide an acceptance test plan with all the items to test. The conversations generated during these tests help adjust the knowledge articles, spoken phrases, variables and tags if necessary. Once the acceptance phase has been completed, the client validates the bot’s release and redirects customer calls to the bot’s number.

Callbot Management and Development

Once the bot has been configured, we will train you to use our software. The aim is to ensure that you can manage your bot (dashboard and analytics), improve its understanding (adding phrases and matching groups) and update its spoken phrases autonomously. You can listen back to user conversations and read them since they are also transcribed into text.

Our teams will provide support for another month after your callbot’s go-live and help you analyse your analytics and past dialogues via weekly meetings. You can also benefit from extended support with a Customer Success Manager if you wish.

For a successful callbot and ROI that meets your expectations, make sure to choose a publisher with real expertise, which provides support tailored to your organisation and training for your staff, to ensure that they fully adopt the solution. 

Feel free to contact us if you would a callbot demo.

There are a lot of players on the French and global bot market (chatbots, callbots, voicebots). There are more or less recent pure-players. More and more, various big software editors add conversational components to their solutions. There are also integrators who offer “tailor-made” solutions.

But not all products are equal. Below is an overview to help you find the best fit for you, depending on your objectives and user experience requirements. This article illustrates the richness of our own conversational strategy.It also provides some guidelines for studying the various solutions on offer.

NLP, the Essential Engine for an AI Bot

Dydu has always had its own linguistic component, which means that we don’t rely on an external player for natural language processing. This enables us to be completely independent. Our linguistic component has a particularly good handle on French and can manage ten or so different languages.

A Simplified Explanation

Our linguistic component calculates the distance between the user’s request and all the phrases in the knowledge base. The knowledge base gathers all the chatbot’s questions, answers and alternative phrases. When this distance (difference) is low, the chatbot can provide an answer to the question. This distance is calculated with a matching score. Depending on the score, the bot either replies immediately, suggests another way to phrase the question, or informs the user that it doesn’t understand. This provides far more relevant answers than a keywords-based system.

This mode of operation also makes it possible to quickly identify any requests what needs to be added to the knowledge base, without having to constantly check the rest. In order to avoid a combinatorial explosion and to minimise the user’s workload, the software has a tree structure of possible phrases, facilitated by matching groups (groups of synonymous or similar words or expressions).

More Specifically

Dydu’s NLP corrects spelling mistakes, identifies lemmas, spots synonyms and hypernyms, and “stems” words. It calculates the distance between all the words to understand the sentence’s general meaning. It divides the user’s request into intents and entities. This enables it to match the user’s questions to content that already exists in the knowledge base, and to answer the overall question. This avoids intermediate choices at each stage and the risk of wrong answers.

This algorithm is at the heart of our solution’s performance, but a number of easy-to-manage features also help to build a real conversational strategy. They all contribute to making the conversation as natural as possible.

Must-Have Conversational Bot Capabilities

• Providing a series of answers on the same subject
• Suggesting alternative answers
• Explaining
• Anticipating the user’s question
• Asking the right questions to provide the right answer
• Clarifying an answer with additional information or a suggested action
• Providing personalised answers
• Knowing how to handle digressions
• Guessing an answer based on the previous questions
• Being polite

Providing a Series of Answers on the Same Subject

In a real-life conversation, you don’t have to constantly repeat the initial subject. This may seem obvious, but a lot of language recognition engines need the topic to be explicit in every question. Dydu’s software manages “bounce conditions” which allows users to ask several questions on the same topic. The user doesn’t have to repeat what their question is about each time. For example, if a user asks: “is health insurance mandatory?”, the answer will be “yes […]”. The next question will probably be: “how much does it cost?”. The bot will understand “how much does health insurance cost”, without the user having to specify it again.

Suggesting Alternative Answers

If a user asks the same question twice, it’s usually because the first answer wasn’t satisfactory. Being able to answer again using an alternative sentence, an image or a practical example particularly useful in a conversation.

Explaining

Some answers are complicated. Just like in real life, sometimes a diagram is better than a long speech. An image or video can be displayed in an open sidebar as part of the answer. You can also provide answers in the form of step actions. This is particularly useful for questions on how to proceed with something.

Anticipating the User’s Question

People often spontaneously finish each other’s sentences, not to interrupt them but to show that they’re listening and interested in what the other person is saying.

The matching feature begins to work as soon as the user starts typing in the bot’s text box. The bot can use an autosuggest feature to dynamically suggest pre-exiting questions. This feature is more sophisticated than autocomplete which is on most mobile phones. You can configure the number and way these suggestions are displayed to optimise conversations. The user can then simply click on their question instead of having to type it in full. This mechanism is based on dydu’s NLP engine. Meaning, once again, that it does not use keywords. This allows to make suggestions using the brand’s vocabulary, even though the user initially typed their own words. For example “customer account” vs “online agency”.

Asking the Right Questions to Provide the Right Answer

In some cases, the bot might need to retrieve certain information to provide the right answer. After retrieving all information spontaneously provided by the user during the conversation, the chatbot will dynamically ask for any missing information (slot filling). This applies for example to a ticket or event search, when it’s important to know the place(s), date(s), time(s) etc.

Clarifying an Answer with Additional Information or a Suggested Action

Depending on the information provided during the conversation with the end-user, it can be useful and relevant to provide additional information. The bot can recognise behaviour and predefined expressions used during the conversation and trigger a specific action. For example, the bot could give an answer and then provide a redirection URL. Or, to give another example, you could set up an automatic escalation to a human operator if the bot is unable to provide an answer more than 2 times during a conversation, or if the words “cancel” or “change my contract” have been used. Each bot administrator can decide what makes the most sense for their company.

Providing Personalised Answers

This contributes to your bot’s perceived value and is probably the area that the dydu engine performs the best. The connection via API with other information systems enables the bot to enrich the dialogue with user-specific information. If the bot is connected to a CRM, it can identify an existing customer with specific promotional codes, for example. For banking, the bot could provide the user with their bank balance, after authentication. If the bot is connected to a HRIS, an employee could check their exact number of leave days and not just how many days off company employees get.

Bots work best when they provide short and precise answers, so you could provide an answer in the form of a decision tree.

Decision trees allow to put the user’s question into context. The bot will ask a series of sub-questions to provide a precise final answer. For example, in France, if you ask the council chatbot about obtaining a civil status certificate, it will ask you to choose between a birth, death and marriage certificate. It will then ask if the person requesting the certificate was born in France or abroad. Only then will the bot explain how to obtain a birth certificate for a French citizen.

Understanding the user context is key to provide a specific and non-generic answer.

Using variables enables the chatbot to retrieve user data during the conversation. And to use it to provide an answer.

These variables can be retrieved from an explicit user statement. Or a transfer of the chatbot’s context (via its webpage, geolocation or telephone number in the case of a callbot). Or data from an API. It could be the user’s name, age, type of contract or any other information that the user may have provided during the conversation, or that has been retrieved via authentication (and therefore accepted by the user within the framework of GDPR).

Knowing How to Handle Digressions

When a conversation “enters” a decision tree, the user can still ask a question on another topic. The bot can answer that question and roll out the end of the decision tree afterwards.

Here is an example:

• User :  “I want to receive my invoices by email.”
• Bot :  “To subscribe to our paperless invoice service, I need your customer reference number. Could you please enter it here ?” (The decision tree is expecting the user to type their customer reference number, but instead they call on another knowledge article)
• User : “Where can I find my customer reference number?”
• Bot :  “You can find your customer reference number at the top of your invoices, to the right.”
• User : “1234ABCD”

And so, the initial decision tree continues…

Guessing an Answer Based on Past Questions

Another of our native features allows to better understand the user’s question without creating a decision tree, based on the questions already asked.

Imagine a HR knowledge base includes the following articles :

• What’s the XX app for?
• How do I install the XX app?
• What’s the YY app for?
• How do I install the YY app?

The conversation might look something like this :

If the engine only takes into account the last question (How do I install it?) and does not look at the context, it cannot determine whether the knowledge article about installing the XX or YY app is needed. But by taking past questions into account, the bot can provide the user with an answer to the question “How do I install the XX app?”.

Being Polite

“Hello”, “goodbye”, “thank you”, “tell me a joke”… Small talk is a key part of conversational experience. Not all bots come with a standard social base, which takes a very long time to create from scratch.

To recap, dydu’s technology is able to contextualise and discuss several topics at the same time, which optimises your user interactions. And that’s exactly what users are looking for: precise, quick and correct answers to their questions, however they ask them. At dydu, we are continually improving our technology and always put ourselves in the user’s shoes !

Withstanding the Test of Time

Chatbots were higly innovative 10 years ago and now feature on most major brands’ websites. They can assist customers with instant answers to simple and precise questions. They were initially implemented by marketing/customer and innovation departments but have since spread to all areas of large companies. Chatbots can therefore also assist employees by answering IT, HR or legal questions. The current health crisis has reinforced 78% of decision-makers’ belief that the advent of bots is inevitable*. This has also been confirmed by analysts (+37% annual growth over the next 4 years). Physical communication spaces have recently been reduced to a bare minimum, forcing companies to find new ways to stay connected. Technology has proved to be a wonderful ally. Remote working has led to a sharp increase in collaboration tools, as well as any tools that maintain an essential link with customers and employees. 7 out of 10 HR professionals also report that chatbots are even more useful in times of crisis**.

Increasingly Visible Chatbots

Chatbots are present on more and more digital channels (websites, mobile apps, instant messaging, social media, etc.). They have become an everyday tool, aimed at more and more users, whether consumers (from sales advice to after-sales service), B2B companies, public service users, start-ups via the NOA chatbot which helps navigate between 10 different administrations, schoolchildren via the CNED’s Jules chatbot, and also employees who need IT advice or want to check their leave balance… In other words, everyone! As such, chatbots have come to represent brands or institutions. Conversational robots are the symbol of digital transformation and a vector of the owner’s image. In fact, 3 out 4 decision-makers  believe that chatbots enhance a company’s image, as a result of their customisable user interfaces and ways of speaking. And 78% of HR decision-makers think that they are an asset for the employer brand**.

Proven Benefits

Far from being a simple gadget, chatbots are real tools for marketing, business, customer services, HR, etc. Conversational robots are inexhaustible and available 24/7. The best natural language recognition technologies, which are becoming more and more sophisticated, can manage bounces (not repeating the subject of a question several times), typos or spelling mistakes, abbreviations, and acronyms, in lots of different languages… They can hold genuine conversations. 80% of the volume of questions asked, represents 20% of the most frequent requests. Chatbots can answer in a relevant and coherent manner… without ever getting angry! As a result, user satisfaction is significantly increased, operations are optimised, and call centre agents or teams can focus on more complex, value-added tasks. Companies thus gain in image and financial efficiency by impacting turnover and operational costs. The initial objectives and measurement indicators should be clearly defined beforehand.

Increasingly Sophisticated Use Cases

Integrating chatbots with company information systems (CRM, HRIS, etc.) and new technologies, helps to finetune services by personalising answers or automated tasks (tracking an order , for example), and also to complete more complex actions directly via the chatbot, such as checking an account balance, electricity or gas consumption, placing an order, booking leave, etc.

What Does the Future Hold?

Tomorrow, chatbots will be able to answer an even greater diversity of requests in an automatic, instant, and natural manner. At the last Vivatch exhibition, visitors could already ask Pepper about the French football league via Botaldo, L’Equipe’s chatbot, to find out who the top goal scorer was or their favourite team’s ranking. In the future, in addition to the current channels, one could imagine asking a robot for directions to a boarding gate at the airport, or a house about that month’s electricity consumption. Beyond these other channels, voice can also clearly complete a conversational robot’s services, to answer questions asked by anyone, at any time, for greater accessibility.

A conversational robot is therefore a real business tool that impacts image, operations, and profitability. As with any project, it is important to define the needs, expectations, objectives, and target. This helps to properly scale the solution to be implemented. It can range from something very simple, with a few predefined questions / answers, to knowledge bases made up of thousands of interconnected items with several webservices, on a multitude of different channels… Thinking bot therefore opens up an ocean of possibilities, with a tool that has already proved its worth.

* Chatbot Observatory – December 2018 

** Chatbot Observatory – April 2020

Over the past few years, B2C companies have moved towards automated customer service solutions, such as chatbots. Their goal? To improve customer experience, increase productivity and reduce costs. And today, they’re taking things one step further by automating part of their calls with conversational voice robots, aka callbots, accessible via a telephone number.

From a technical viewpoint, a callbot functions like a chatbot, but with two extra components – voice recognition and voice synthesis. This means that they can understand a customer or lead’s request in natural language, give an answer, and even perform certain tasks. Callbots are generally deployed by companies with a high volume of calls (insurance industry, telecom, transport, e-commerce, hospitality, etc.).  

The next part of this article presents 3 callbot use cases that help effectively reduce pressure on customer service teams, illustrated by real-life examples.

Get Information and Book an Appointment

Just like chatbots, callbots can provide general information: opening times, address, information about services and products, etc. If the user is an existing customer and wants to identify themselves, they can benefit from personalised information based on their customer account. There are several ways to identify a user: via their telephone number, with their first and last name and/or customer account number, etc. The company puts in place the script and login variables. 

If the conversational bot is connected to the company’s booking management software, it can also suggest that the user make an appointment and pre-qualify the request (context, nature of appointment, etc.).

Sign up to a Contract

Insurance or electricity contracts, Internet packages… Callbots are excellent ways of increasing the number of new contracts, especially during non-working hours.

How does it work?

After asking the user to identify themselves, the callbot asks a series of sub-questions that are needed to create the contract. In the insurance sector, for example, these questions will relate to the type of insurance (car, house, civil liability, etc.), vehicle, make, model, contract, etc. The callbot pre-qualifies and records the request. Unlike IVR, the user doesn’t select options with the keys on their telephone, but answers out loud in their own words or those suggested by the bot.

The callbot then informs the user of the next steps and which documents they need to provide in order to complete their request. At the end of the conversation, the user receives a confirmation message (SMS and/or email). Depending on the insurance company’s strategy, the new customer may be called back the next day by an agent to confirm that their new contract is in place. If, however, a customer wishes to terminate their contract, the callbot may hand over to a human operator.

Manage Complaints or Make a Claim

One of the most frequently contacted departments in a company is the after-sales service. The latter should be perceived as a real strategic lever for customer satisfaction and loyalty. Automating part of this after-sales service can be particularly effective in making the processing of requests smoother, and therefore increasing customer satisfaction.

Managing complaints is a frequent callbot use case. Specific scripts with advanced decision trees can be put in place, to personalise information provided to customers as much as possible, as well as information given to the company. For example, the callbot could ask unhappy customers if their call is about a current request or a new claim, and run a script based on the answer.

• request in progress: identifies the claim with a tracking number, gives an answer according to the request status (in progress, processed, resolved, failed…), as well as any potential actions to be carried out: upload a form, send supporting documents, etc.
• new claim: asks sub-questions to collect all the information required to handle the problem (type and frequency of breakdown, type of product…), retrieves technical information and collects ID data in order to finalise the claim.

Messages should be fully customisable and modifiable by the company at any time. As with breakdown or incident reports, insurers can also set up callbot scripts for claims. The principle is the same, only the callbot’s questions are different: date of claim, context, registration number or postcode… The bot can also offer telephone assistance if necessary.

To Take Things Further

Whatever the case, a callbot can record conversations and replay them later (the user must be informed beforehand). This enables the bot’s manager to continually improve it, enrich the CRM with additional information and better understand the customers’ needs and friction points. The bot can also transcribe the conversation in order to leave a note in the CRM, for example.

If connected to third-party technologies, such as RPA (Robotic Process Automation), the callbot can automate tasks such as creating a contract, filling out a claim report, updating customer account information, etc. directly in the company’s software (CRM, IS, ticketing tool, etc.). Once created or modified, the documents can automatically be sent to the customer for signature. This allows an end-to-end automation of simple processes, saving both the customer and supplier time and resources.

The callbot can be configured to optimise the brand’s sales strategy as much as possible, by increasing sign-ups, monitoring, and managing claims or termination requests… The tool’s profitability is therefore easy to calculate.

If you want your users to interact with your chatbot and, most importantly, to keep coming back, you need to provide a quality service, with a bot that understands user questions and provides the right answers. Beyond the product itself, you can also implement other good practices to ensure your employees or customers instinctively reach out to your bot, starting from the earliest stages of development.

Involving End-Users

Our first best practice is to involve all stakeholders in your chatbot development. The people working on the project are the first to come to mind, but we also recommend putting together a panel of target users. This could include customers for an external bot, or employees for an internal one (IT, HR, legal, etc.). You can get them involved in defining the bot’s identity (choosing a name, the colours, avatar, tone, etc.), and ask them for feedback throughout the implementation stage.

Why is it a good idea to involve users in the creation of your bot? Firstly, to ensure that it meets your target audience’s expectations in terms of objectives, content (scope and knowledge), identity, and channel. A panel of test users will enable you to identify personas, and the use cases to address. Users will also feel more engaged if they are involved in the creative process. They will naturally become ambassadors of your new tool and raise awareness within the company.

Let’s Take Some Examples…

When designing their HR chatbot, the Conseil Départemental du Val d’Oise tested their first knowledge articles on a group of pilot employees and managers. They then finetuned this knowledge and defined the bot’s behaviour by personalising its identity and ways of answering. “An internal group of users selected the avatar and chatbox colours, and we collected HR employee feedback. Involving them in the process has led to greater ownership of the tool,” Ophélie Lannoye explains, support officer for cross-departmental HR projects at the Conseil Départemental du Val d’Oise.

As for Société Générale, they made a survey to collect ideas for their internal virtual assistant’s name.

Another example is Noa, the conversational robot created by the Prefecture of Paris and Ile-de-France. Noa is an expert chatbot in administrative procedures relating to the creation of a company. During the creative phase, the project team organised several workshops with entrepreneurs. They compiled their needs and expectations, and collected feedback about the bot and its first knowledge articles (content in the form of questions/answers).

Maximising Bot Visibility

If you want users to interact with your bot, it needs to be visible and easily recognisable:

• deploy it on all of your extranet or intranet pages
• opt for colours that contrast with the background
• make the bot button visible. Users should immediately know that it’s a bot, or at least an online help service. You could also add a little phrase, such as “need help?”, next to this button.

Our client’s home page, City-Drop

You can also “trigger” your bot to encourage visitors to use it, after several seconds of website inactivity for example. The CNIL recommends not opening a chatbox pop-up window before obtaining the user’s consent. But there are other ways to catch the user’s attention and encourage them to click on the button, such as making the avatar move, or displaying the “need help” phrase.

Guiding Users to Avoid Disappointment

Chatbots are not supposed to be able to answer everything, the scope and themes are defined beforehand. It is important that the bot informs the users of its areas of expertise to avoid any disappointment. This can be in the form of an onboarding message before the conversation begins. The message should list the subjects the bot handles and give users some tips on how to address it so as to maximise its understanding.

For the first few weeks after deploying the bot, we also recommend displaying an information banner, asking users to be lenient if the bot doesn’t understand everything. You can also use this message to explain that the bot will improve with practice and will soon be able to understand even more questions and phrases.

Displaying a “top 3” of the most frequently asked questions can also help users identify what the bot can help with, and easily understand how to phrase their questions.

Finally, when the user has already opened the chatbox, you can trigger a knowledge article to encourage them to continue their conversation with the bot:

• “Would you like any more information about our new offers?”
• “Feel free to ask me another question, I’m happy to help ”
• “You currently have 1 item in your basket, would you like to see our new products?”

Communicating

We have already discussed the benefits of communicating before the launch of your bot. This could be a survey on social media to help pick the future bot’s name, or an avatar vote. Before implementing their internal bot, Société Générale distributed chatbot-themed goodies to their employees, to get them interested.

As with any web project or new communication channel, it is important to implement a communication plan from the moment the bot is launched. We strongly recommend promoting your new tool via existing communication channels, such as a newsletter, social media post, blog article, intranet or website message, internal posters, or even a video!

Video for our client Sosh’s online “Help and Contact” tool

To celebrate the launch of Noa, the Paris Prefecture organised a party at Station F with all the participating public administrations, entrepreneurs from the creative workshops, and all the start-ups at Station F. They also presented the bot at the Vivatech show. Noa the chatbot even has its own Twitter account!

Analysing, Testing, Improving

Once the bot has been deployed, there is still lots to do! It is important to monitor the bot’s performance indicators, in order to improve it. The number of dialogues and interactions, matching and satisfaction rates, analysis of failed dialogues and misunderstood sentences… These will all help you improve your content by adding, deleting, and grouping questions, rephrases and synonyms.

This analysis should be carried out regularly (we recommend that you spend 0,5 to 2 days per month on bot management). But it is especially important in the days and weeks following the launch, because the best way to encourage customer and employee bot adoption, is quality.

Pixelraise partnered up with dydu to implement a gamification layer to the Jules chatbot, a homework helper for secondary school students, deployed by the CNED with dydu’s solution. Moezz Sedkaoui, founder and managing director of Pixelraise, looks back at this collaboration and explains the advantages and mechanics of a chatbot-gamification association.

Can you describe your company and role in a few words?

Pixelraise is an IT consultancy firm and web and mobile developer, with an entity in France and Switzerland (WebGoeSon). Our 25 employees develop custom IT solutions, e-commerce sites, native and cross platform mobile apps, and CMS-based sites. Pixelraise has a large client portfolio: Merck, EMS Medical, Produits Dentaires, Carglass Suisse, the French Ministry of Education, Sword, Hermes, Plezi, etc.

How was the Pixelraise and dydu partnership born?

Dydu put out a call for tender looking for a partner that could quickly gain the skills required for their chatbot technology and, above all, enhance the existing Jules chatbot app through gamification. A shared contact brought the decision makers from both companies together, giving rise to a great collaboration and successful project over a 10 month period.

Can you describe the role Pixelraise played in the CNED Jules project?

Pixelraise was responsible for developing the gamification layer (excluding UX and design), with 3 main objectives:

• develop digital uses for the “Devoirs faits” programme (personalised support and help with homework for secondary school students)
• reward effort with prizes, to encourage students to use the app 
• motivate users to play regularly

What does the Jules chatbot currently offer in terms of gamification?

The gameplay we developed enables the user to win flames and gifts. By earning flames, the user can move to the next level, and is rewarded with a gift at every level. These gifts include outfits for the user’s avatar or backgrounds to personalise the environment.  

What were the different steps to implement gamification in the chatbot? How does this combination work technically?

The first step was to understand and perfectly master the dydu component (Assembly.js), the chatbot’s engine that contains the knowledge base. Then, we developed all the APIs to interact with user actions, to reward and encourage them to learn while playing. Technically, the interactions between the quizzes and dydu’s chatbot are done via a function call in Js, using JavaScript Window Option. Finally, the app authentication is based on an API managing the creation of user accounts (student and teacher profiles – web and mobile) and the connection.

What are the main game mechanics and keys to a successful game or challenge?

Gameplay is the first aspect, including playability and the rules of the game. The player should be immediately immersed in the action with an adapted design. The second important aspect is finding the right balance between constraints and possible actions; too many rules make the game difficult to play and too much freedom leads users to drop out, without any retention. The last aspect is the pleasure or fun of playing, based on innovation, whether technical, design or narrative. 

The keys to a successful game can be summed up in 5 points:

• choose a unique gameplay; use well-known game mechanics that the majority of gamers are familiar with, and add customisation to make the game unique,
• opt for an innovative design; UX is very important, and it is essential to carefully design the game so as to offer a unique and coherent experience from the very first session. The game should be fun. If the user’s first impression is negative, there is little chance they will play again (and they will probably uninstall the app),
• work on retention; for example, create addiction with random rewards,
• optimise the response time; make sure that your game takes no more than 2.5 seconds to load, in order to reduce your abandonment / unsubscribe rate, 
• market the game with a suitable budget to make it visible among the many games available, in a very competitive market.

What other web and mobile developments have you carried out for the Jules project? (application, authentication interfaces, etc.)

We developed an external back-office on the Java / Springboot / BD Postgress / J2ee stack, in order to manage certain items such as the flames, quizzes, levels, outfits, backgrounds, etc. It also contains user information.

Do you think there will be other opportunities for dydu and Pixelraise to work together, beyond the Jules project?

Absolutely, we are complementary and perfectly capable of integrating dydu’s chatbot solutions. We recommend dydu to our clients as an expert partner in chatbot solutions.

In your opinion, how is UX design evolving in web projects?

Over the past 10 years, the user experience (UX Design) has taken a new direction, and users are now at the heart of projects (User Centric). UX design is not just about the design aspect of the interface. It also involves a reflection on the standard user journey in order to meet user expectations. The science behind the user experience is central to all web projects because more then 7 out of 10 internet users will not return to a website if their first experience does not live up to their expectations.
P.S.: by the way, I’ll soon be announcing the release of our UX addict websites

How do you think Artificial Intelligence will impact UX?

I believe that artificial intelligence will revolutionise the user experience. Artificial intelligence and user experience are closely linked in today’s world, where we are now facing a turning point in a new era including big data, voice search, chatbots, etc.

Real AI allows a degree of customisation that is hard to find on other more traditional channels. Within the context of voice search in particular, AI makes it possible to introduce greater personalisation, fluidity, and intuitiveness, all of which greatly benefit the UX. It is important that UX designers offer three options at the most per interaction, because the user will struggle to retain more than three.

Conversational technology is the easiest way to access information. Dydu was founded on this belief 10 years ago and it continues to be true today. Bringing technology closer to man as opposed to forcing man to adapt to technology has finally become a priority. But what may seem obvious in theory, is more complicated and labour-intensive in reality, especially if the aim is to reach each individual, with their own specific characteristics.

The Regulations

There are a lot of regulations surrounding digital accessibility, which consists in making online services accessible to people with disabilities. In France, General Accessibility Guidelines (RGAA) are issued by the Interministerial Digital Directorate (DINUM), the fourth and latest version of which were published in October 2020. 

These guidelines do not apply to all organisations and companies. For some, the regulations are compulsory, such as digital public services and, more broadly, communication services for the public managed by legal entities under public law (local authorities, etc.), private companies delegated with public service or general interest missions, private companies with public sector participation, and companies with an annual revenue of more than 250 million euros. For others, digital accessibility may be a part of their inclusion strategy, as more and more companies are incorporating these guidelines into their digital strategy.

Chatbot Accessibility

A lot of dydu’s clients are required by law to comply with the RGAA standards and are regularly audited. For dydu, beyond the regulations, this approach is a priority to ensure the coherence of our products and the values of accessibility we have developed since day one. Although compliance with the RGAA regulations is our client’s responsibility, we make sure to incorporate this aspect into our solution and to provide advice when needed. 

A bot is made up of 3 parts: the motor (natural language processing algorithm), the content consisting of the knowledge base, and the user interface part that is the chatbox. Digital accessibility applies to the last two items. Our solution ensures that our clients are in full capacity to comply.

As such, we have integrated the RGAA standards that apply to dydu as a content editor (WYSIWYG) into our Bot Management System, which covers images, lists, attachments, anchors, bold formatting, iframes, links and tables.

The design of our dydubox – a default chatbox that can be customised without code by our clients – also takes into account the RGAA guidelines and ensures that our clients are fully compliant with the rules. This is the case for headers, footers, titles, buttons, visible links, the navigation, generic chatbox content (satisfaction bubbles, GDPR text, etc.) and the selection of the language used. Our design is compliant, but our clients can modify it via the dydubox interface and can therefore move away from the RGAA criteria if they so wish. Dydu has therefore created a document to list the areas requiring particular attention and the rules to apply in order to remain fully compliant.

Dydu’s goal is to enable bot administrators and managers to configure RGAA-compliant content automatically, without changing the source code, while maintaining the end-user experience. Users should be able to easily obtain a bot’s content or locate possibilities to discover content that is better suited to their situation, such as the description of an image.

These regulations don’t apply to all organisations. Administrations, public and para-public sector organisations, as well as all large companies are bound to comply, and although in theory everyone should, it can be difficult for smaller organisations. The final decision lies with our clients. We do everything in our power to ensure that they can be fully compliant. Digital accessibility is at the heart of our values and our product strategy.

What if your shop was open 24/7? Your website is permanently available, but – fortunately for your employees -, your in-store or telephone advisors are not.

One of the reasons customers appreciate direct interactions or contact with real-life customer advisors is because they can ask specific questions and be guided in their decisions. 73% of consumers identify customer experience as an important factor in their purchasing decision (PwC). With the current health crisis showing no signs of slowing, more and more consumers are turning to online shopping. Their expectations, however, remain the same. Chatbots can be an excellent compromise as they can provide part of the expected experience at any time, on any device – whether mobile or fixed -, via any channel – brand website or app, instant messaging, etc.

To understand how, let’s take a look at a customer service advisor’s different roles…

Welcoming

Just like a front of store attendant, you can position a chatbot at the entrance of your brand’s digital space, to embody your visual and behavioural communication style. Your chatbot could be very visible and forward, immediately striking up a conversation, or discreet and only available on request via Push rules or behavioural targeting. The latter involves observing an end-user’s browser behaviour to trigger your chatbot at exactly the right time. For example, if a user is inactive for several minutes, your chatbot could pop up and offer to help.

Whatever you decide, the way your chatbot interacts with your leads and customers should be in line with your corporate identity. Chatbots are pleasant and considerate thanks to the welcome phrases and/or information banners adapted to the time of year (Christmas or other holidays) or your business news. Chatbots are also always be polite thanks to a social base, that includes nearly 700 knowledge articles or “small talk” phrases. And finally, chatbots are, by definition, patient, as they wait for
questions and can even pre-empt them by displaying Top Knowledge or most frequently asked questions; either suggested by the algorithm or hand-picked from those highlighted in your analytics. They are always even-tempered, and you can “let them think” via an optional loader so as to not display the answer right away, contributing to the many benefits of natural language recognition that facilitate man-machine dialogues.

Guiding

A customer advisor’s main role is to help visitors make a choice. Chatbots can respond to an end-user’s specific needs, instead of simply providing a generic answer, using a number of contextualisation features. Decision trees allow to formulate sub-questions in order to reach the final answer. Variables enable the chatbot to retrieve user data during a dialogue and use it to answer the customer’s question. Slot filling requests any information needed to make a recommendation – for example, a date and destination for a train ticket, or hotel booking…

Displaying

As any good salesperson knows, understanding the customer’s needs is key to offering them the right product or service. Chatbots can highlight the right product or service at the right time…

If chatbots make you think of little text bubbles, it’s worth knowing that images and videos can also be added to a gallery, and then displayed when appropriate in the side panels, thereby showcasing certain products and making them more visible. A nice photo or carousel, for example, is always more enticing than a descriptive text.

Assisting the Purchasing Process

Leads on the verge of becoming customers sometimes need help to complete their transaction. How do I pay, how does delivery work, how do I return products, etc. All these questions are important because they are likely to discourage customers if they’re not well informed. This stage often requires more explanation and clear step by step assistance, which is where “step actions” can be particularly useful.

You may prefer to redirect customers to real-life agents at this crucial moment. Additional answers allow you to configure features such as the use of keywords – signing up to, changing or terminating a contract, for example – or successive non- responses, to automatically switch to a human operator.

Chatbots or callbots are not intended to replace your teams, but to complement them. Implementing the right functionalities to meet specific objectives will contribute to your tool’s performance. The better trained and adapted to the situation your chatbot is, the better it will help you convert your leads and satisfy your customers.

A Growing Use Case

The way chatbots are perceived by Human Resources has rapidly and positively changed over the last 2 years, as shown by the Chatbot Observatory’s last study. Recruitment, in particular, is an area in which the increase in expectation is the strongest. It constitutes a use case in itself for 40% of respondents (+12pts in 2 years), and 63% of decision-makers believe that chatbots will become an essential tool in the recruitment process.

Recruitment is very much a two-way seduction process, and chatbots impact both the brand’s image as an employer, and also the applicant’s experience. Chatbots are a vector of innovation for all parties involved and positively contribute to the brand’s image. Moreover, a chatbot portrays the employer in the best possible light, by providing applicants with a dedicated service 24/7, immediate and relevant answers, encouraging them in natural language and helping them with their job search. Particular attention is paid to the candidate’s experience and background, saving them time and enabling them to finetune their application to the company’s needs
and expectations.

Meanwhile, employees also feel more valued, as they can focus on meeting and interviewing new talent, rather than on the repetitive and administrative aspect of managing applications.

Adapted to all Profiles and Generations

Means of communication vary depending on the type of profile and generation. Chatbots can be deployed on lots of different channels, such as websites, mobile apps, instant messaging tools, or social media, enabling employers to reach out to potential hires, particularly the youngest on the job market. Natural language recognition means that everyone can express themselves simply and, combined with other tools, contributes to an unbiased recruitment policy favouring diversity.

Key Features

Several features can help improve the applicant’s experience. Being able to use them could therefore prove decisive when choosing the chatbot platform! Your recruitment needs will change over time. The number and type of jobs will vary. You can immediately attract and direct applicants to key job availabilities via an information banner, to focus their attention on your priorities.

Applicants want to save time and feel that they are being taken care of. The top knowledge feature can be used to anticipate the most frequently asked questions or highlight any relevant information. This can be positioned at the top of the chatbox, either manually (you choose the knowledge articles
to include), or automatically (most frequently asked questions), so that the user sometimes doesn’t even need to type their question. If the company has lots of opportunities to offer, you can add a short presentation video about specific jobs or sites, that can be accessed via the side panels. This is also a great way to present the company’s specific recruitment process through tutorials or step-by-step actions, using the Step actions feature.

In addition to delivering generic information, which is often very useful, you can also personalise your answers. If the chatbot is connected to the company’s system via an open API, for example, the applicant can track their own application. By typing their file number into the bot, they can obtain information about its status. And finally, applicants can also use the bot to book a meeting if they get through the first selection phase…

Recruitment chatbots should therefore be a key part of your strategy, as vectors of a positive image and a genuine service for applicants. Designed to meet HR operational requirements and also promote the brand’s image, they help the company stand out and attract the best talent.

You’re interested in implementing a chatbot to improve user satisfaction and increase productivity, but you don’t know which software to choose among all the publishers on the market… You’re concerned about overly limited FAQs, or chatbots that don’t understand what’s being asked of them… Sound familiar?

The level of success for a conversational robot mainly depends on the quality of content, and the bot’s ability to understand questions and retrieve the right answers from the knowledge base. This article provides advice about which technology and functionalities to prioritise, in order to make your chatbot project a success.

The NLP Algorithm, your Chatbot’s Engine

The technology behind chatbots, called natural language processing, is essential for it to work properly. This technology is based on an NLP algorithm (Natural Language Processing), that identifies the user’s question and matches it with the suitable answer.

There are several natural language understanding methods – distance calculation, syntactic analysis, and keyword matching. To find out more about how natural language processing works and the benefits of each method, read this article. 

All technologies are not equal, and it is important to find out which method is used, the algorithm and matcher’s performance (by testing), the connection with external matchers, and to identify whether the publisher owns the technology or uses a third party software.

Essential Functionalities

• Politeness and Small Talk

Your bot should be able to hold a smooth and natural conversation with your users. Like a human, it should therefore understand and use key phrases (hello, please, thank you, etc.) and small talk (compliments, jokes, etc.). To save time and to avoid creating it yourself, ensure that your bot’s knowledge base includes a social base.

NOA, the Paris and Ile-de-France Préfecture chatbot :

• Formulations

There are lots of different ways to ask the same question. You chatbot needs to be able to understand them all. Matching groups (groups of words and phrases that mean the same thing) and formulations allow to add these different words and phrases under a same knowledge article.

• Activity Monitoring and Improvements

You’ve created and tested your knowledge base and deployed your bot on the channel of your choice. But that’s not the end. It’s important to regularly monitor your bot’s activity, to continually improve it.

There are several functionalities to help you with this. Firstly, analytics, which indicate, among other things, the quality of your interactions and dialogues (the bot’s level of understanding of the questions and its ability to provide the right answers) and user satisfaction with regards to the answer given. Quality alerts are also useful to check the quality of content according to predetermined and customisable criteria (broken links, long answer time, etc. ). Other functionalities can help continually improve your base. Analysis of failed dialogues or misunderstood sentences enables you to add new phrases to an existing knowledge article, or to create a new one. You can also identify similar knowledge articles in your base and choose to group them together or make them more distinguishable. 

Are you interested in creating a chatbot? Feel free to contact us. 

Over the past few years, remote working has become more and more widespread, increasing sharply since the beginning of the health crisis. Along with the many advantages this presents for employees and businesses, a number of new challenges have also emerged. How can you improve efficiency and teamwork when your employees are isolated? How can you ensure that your employees have mastered their working tools? How can you relieve IT departments of their increased workload?

This article will explain how a chatbot solution can help IT departments meet these new challenges, and also take a look at Zen, the first chatbot dedicated to Microsoft 365 created by Ai3 in collaboration with dydu. Ai3 is a consulting company and innovative solutions integrator, specialising in Microsoft solutions. It recently became part of the Talan group.

Chatbots, an essential tool for IT Departments

Over the past few years, more and more large and medium-sized companies have implemented internal chatbots, primarily to relieve pressure on the departments that receive the most requests, such as the Human Resources, IT or legal departments. 

Among dydu’s 160 bots in production, almost 30 are IT/helpdesk chatbots that manage several hundred to several thousand dialogues per month, depending on the client company’s size. IT bots are far more than a passing trend – there are real benefits for both the end users and businesses that implement them.

• An Improved Customer Experience

Password management, software or application support, equipment orders, account management… Chatbots can handle all the simple questions relating to IT and office applications and pass any more complex requests over to the support teams.

Conversational robots provide instant answers, 24/7, on any channel – intranet, desktops, mobile apps, messaging platforms… They can act as an IT department’s single point of contact for questions and are fast and accessible. The answers are all created and approved by the IT department, so they are reliable and consistent, and can also redirect to existing content such as documents, intranet pages or external links, and provide step-by-steps, tutorial videos and explanatory images directly in the chatbox or in the associated sidebar.

IT chatbots are particularly useful because IT requests are often urgent and need to be resolved so that employees can get back to work. This tool not only provides immediate answers, but also 24/7 assistance, which is a real asset, particularly for employees working staggered hours. Virtual assistants help users complete simple tasks by themselves, through existing company documentation – tutorials, troubleshooting guides, step-by-step instructions, etc.

• Optimised Operations

Implementing an IT chatbot significantly reduces the number of incoming tickets and improves the request management process. IT chatbots reply to all the simple and frequent requests, thereby reducing the number of support requests and enabling IT managers to focus on their primary mission – to improve the company’s IT system, thereby improving work efficiency.

By encouraging employees to be more autonomous, the IT department’s productivity is increased and costs relating to support are reduced. Chatbots are particularly effective during IT migrations, which usually generate a lot of questions. They can also provide general information in advance by displaying version upgrades, potential interruptions or future benefits.

Zen, the first chatbot dedicated to Microsoft 365

As a support for Microsoft solutions, Ai3 decided to create a chatbot to answer questions about the use of applications in the Microsoft 365 suite – Excel, Outlook, Teams, OneDrive, SharePoint, Skype, Yammer, Kaizala, etc. Ai3 used dydu’s technology to deploy Zen.

• A Comprehensive and Scalable Knowledge Base

Knowledge articles:  Zen covers all the usual themes, such as messaging, collaborative work and office tools. The knowledge base includes more than 600 knowledge articles that are regularly updated by SoYouz, Ai3’s digital transformation consulting agency. 

Matching groups: The bot also includes matching groups, that group together several ways of saying the same thing (for example, Excel: worksheet, spreadsheet, XLS, Microsoft Excel, etc.). Matching groups facilitate the bot’s understanding of a question. The knowledge articles and matching groups are based on more than fifteen tools from the Office 365 suite. 

The social base: In addition to IT knowledge, the bot has a social base, meaning it recognises forms of address and ways of speaking that make the dialogue smooth and natural. This applies to all dydu bots.

All our company clients share the same knowledge base, meaning that their chatbots are always up to date, without any actions required on the client’s part. Dialogues are monitored, and when a knowledge article is added, it is added for all clients. It is however possible to personalise about ten specific knowledge articles (IT support contacts details for example).

All knowledge articles are in both French and English. It’s easy to switch from one language to another with a button in the chatbox.

• A Turnkey Solution

Chatbots are very easy to set up – companies are provided with a code to deploy the chatbot on the channel of their choice (webpage, intranet, support site, etc.). All the client has to do is personalise ten or so knowledge articles and integrate the code on their chosen channel(s).

• Advanced Features

Chatbots answer questions in a text format, but they can also provide rich content, such as links, images, GIFs, step-by-steps or videos in the chatbox or sidebar for better readability. They can also ask other questions on the same subject to provide further information. If the bot misunderstands a request, it calculates the probability of the question and suggests reworded alternatives.

Want to know more about the Zen chatbot? Feel free to reach our team !

How can you improve your customer experience and strengthen support without increasing the workload for your customer services? By encouraging customers to be autonomous with selfcare solutions!

FAQs, chatbots, community forums… Over the past few years, these tools have become part of customer services, enabling users to find an answer to their question quickly, and to solve problems themselves. Selfcare solutions can reduce the number of incoming requests by 30%, thereby improving your customer service department’s productivity and lowering costs. Meanwhile your customer experience is also improved because the customer’s question is answered quicker than it normally would be, exactly when they need it, even late at night or during the weekend.

Certain tools, such as chatbots, by directly performing actions. This is made possible by several methods, including the combination of chatbots and RPA (Robotic Process Automation).

This article explores the benefits of this solution for customer services, particularly in the tourism industry.

Chatbots and RPA – from information to action

Chatbots can provide users with generic answers, contextualised ones (depending on whether the user is using a computer or telephone, their location, or other variables), or even personalised ones (if they have been authenticated).

To enable customers to perform actions from the bot, such as checking their latest bill, filing an insurance claim, editing information on their customer account, or changing a reservation, the company needs to connect the chatbot to its tools (CRM, IS, reservation management tool, etc.). Several methods can be used for this:

• Connecting the bot and software via a dedicated API (Application Programming Interface)
• Coupling the chatbot with RPA (Robotic Process Automation) technology, which is also connected to the company’s various tools

“RPA is the technology that allows anyone today to configure computer software, or a “robot” to emulate and integrate the actions of a human interacting within digital systems to execute a business process. RPA robots utilize the user interface to capture data and manipulate applications just like humans do. They interpret, trigger responses and communicate with other systems in order to perform a wide variety of repetitive tasks. Only substantially better, an RPA software robot never sleeps, makes no mistakes and costs a lot less than an employee.” UI Path 

Chatbots understand the user’s request and simulate human conversations, whereas RPA automates collection, analysis and information writing operations, in order to edit documents, forms and web pages without human intervention, even if human supervision remains important.

The benefits of automating processes via a chatbot

The main benefit for users is the availability of the tool and the time it saves. Chatbots handle the entire request immediately, even outside working hours. As for companies, they can extend and refocus their human operators’ productivity, generate lower costs, and reduce the risk of errors. 

The advantage of RPA technology is that it serves as a single interface for the bot. Companies don’t have to deploy a dedicated API for each application / software they wish to connect to the bot.

Automating processes via a chatbot is beneficial for customer services that receive a lot of requests, such as companies in the banking, insurance, travel, logistics, consumer goods, telecommunications and healthcare sectors.

Use cases in the tourism industry

Since last March, a lot of people have had to change or cancel their reservations. A train ticket, a holiday rental, a plane ticket to go abroad, or any kind of booking for an in-person event… Employees in the tourism sector have therefore had to cancel or change reservations multiple times. Their job consists in answering customer questions, cancelling the initial booking, sending an email confirming the cancellation, and sending a credit note or refund. In the event of a modified booking, they have to reserve a new time or date and send the customer their new reservation details. These repetitive and time-consuming tasks coincide with a difficult economic period for the industry, which has led to partial or total unemployment.

Devoteam, specialising in innovative technology consulting, UI Path, a robotic process automation company, and dydu have combined their expertise to automate reservation modifications and cancellations by associating a chatbot (dydu technology) with RPA (UI Path technology), integrated to the client’s CRM.

A search for the customer’s booking can therefore be carried out using their name and reference number, entered into the chatbot (or any other relevant identifier). The robot then edits the contract, changing the dates or cancelling the reservation, saves the new contract and sends it by email. The chatbot then informs the customer of an incoming email.

This is an unprecedented solution for the tourism industry, enabling players to respond instantly to the growing number of customer requests and relieve pressure on their operational teams. This tool was initially designed for hotels, but could also apply and be extended to restaurants, airports, transport companies, etc. With the aim of offering an ever more qualitative customer experience, while addressing the industry’s current economic and social constraints.

As businesses and individuals are increasingly moving towards new technologies, it is time for public services to adapt too. The objective? To become all-digital by 2022. A complicated yet necessary project, further highlighted by the current Covid-19 crisis. Public service bodies need to be modernised in order to offer a quality service, that can be accessed anywhere, at any time.

However, only 39% of French administrations have reported using digital technology in a substantial way to manage their user relationship*. This transformation is therefore in early stages and must be accelerated so that public services can meet future challenges. This is essential for both users as well as administration and local authority employees.

Multiple challenges

Citizens today are interconnected, and administrations must adapt to this social change. The challenges are multiple – to provide easy access to information, simplify administrative procedures, improve productivity to the lighten employee workload, etc. By changing their mode of operation, public services can rapidly gain in efficiency. Of course, those who stand to benefit the most from this are the users, but unfortunately not everyone is in the same boat. Indeed, this is the main limitation of digital transformation, and the biggest challenge – reaching everyone. Some people have little or no access to digital tools, or to the training they would need to tackle these new processes alone. This part of the population needs support and guidance, to ensure that these procedures are as accessible and easy-to-understand as possible, for everyone.

The French state has begun to move towards a paper-free system through “modernisation” reforms. The next step is the adoption of various new technologies, particularly those based on AI. This article will now focus on the benefits of chatbots and their use in public services.

The role of chatbots in the digital transformation of public services

• Relieving administrations and facilitating internal processes

The French administration is often criticised for its slow pace and limited opening hours. Yet public services are increasingly in demand. They are easily overwhelmed and processing a file that should take a few minutes can sometimes take weeks. Which is the advantage of a conversational robot – a first point of contact to provide answers to simple and recurring user questions, and to escalate to a competent agent (livechat or telephone) if necessary.

“40 to 60% of user requests to local authorities do not require any special skills, such as looking up the opening hours for a swimming pool or the address of a recycling centre,” Yannick Sourisseau explains, journalist and editor-in-chief of Ville Intelligente Mag.

For certain departments, automating the processing of simple and recurring requests could save a considerable amount of time. The relationship between the user and administration would then become smoother and the service provided of a better quality. The entity contacted could then give a clear and efficient answer in a more reasonable timeframe.

When associated with other databases and internal software, chatbots can also facilitate internal processes and user procedures. This is referred to as RPA (Robotic Process Automation) chatbots, enabling requests to be handled with no human intervention, although human supervision remains essential. This could include subscriptions, changing information in a file, editing a document, statements, etc. All saving considerable time for both the user and the agent.

• Providing a continuous, easy to access service

Chatbots can be accessed at any time of the day, every day of the year. This enables public services to smooth the pace of requests and lighten the load at high-traffic hours. Chatbots are available all the time and also on all communication channels – websites, mobile apps or messaging apps. Chatbots favour a transversal way of working between departments and administrations by providing a single point of contact for all their questions.

Users benefit from an immediate, homogenous response, 24/7, and therefore have a better experience and image of the service provided.

• Facilitating administrative procedures and access to information

A chatbot’s main role is to help users find information. A lot of administrations implement bots to make administrative procedures, that can sometimes be daunting and complicated, easier to understand. Take for example Noa (“Nous orienter dans l’Administration”), a chatbot for business owners that answers their questions about administrative procedures when setting up a business – tax and legal obligations, patents, how to register a trademark, financing, recruiting staff, etc.

This conversational robot was implemented by the Île-de-France prefecture, grouping together information and procedures for more than 10 administrations at a single point of contact. This enables business owners to gain a better understanding of the procedures and legal obligations linked to their activity and to become more autonomous and less dependent on an administration’s response time.

Since NOA was implemented in March 2019, business owners have asked more than 6,000 questions, with an 80% accuracy rate. It should also be noted that when lockdown measures were announced, users asked more than double the amount of questions compared to the previous period, with a notable peak during the first 3 weeks.

This chatbot is the perfect example of a transversal tool used by completely separate departments and administrations. As well as providing useful information, this type of chatbot could speed up administrative procedures in the future by carrying out actions directly via the bot, such as submitting a file.

• Encouraging “selfcare”

72% of consumers prefer to find answers to their questions without contacting customer services.** Most customers turn to the internet first, whether via an online search or directly on the company’s website. They expect to be able to find an answer by themselves, in real time, at any time of day or night. Most companies have therefore made content available to their customers, so that they can be more autonomous – tutorials, documentation, guides, FAQs, chatbots, community support platforms, etc. Users are therefore used to using these kind of tools and expect the same from public services.

The lockdown period further highlighted this observation, as it became necessary to find new ways to communicate while people were isolated at home.

For example, schools and universities had to adapt quickly to continue teaching their classes. The national education system has already started providing schools with new ways to teach and communicate, that are more in line with pupils’ habits and behaviour, particularly by developing digital platforms promoting distance learning.

Jules, the virtual homework helper for secondary school students, implemented by the CNED in January 2019, is an example of this. Specialising in French, Maths, History and Geography, Jules is regularly updated by teachers and provides answers to questions about the curriculum from Year 7 (11-12 year olds) to Year 10 (14-15 year olds). It also provides methodology sheets, advice, quizzes, tips and videos. During lockdown, the chatbot experienced a peak in traffic with nearly 235,000 interactions per month, and currently has nearly 70,000 users. This provides valuable and free support for students, making them more autonomous and motivated.

Chatbots can also be useful to help those who are digitally illiterate because they process natural language.

• Making everyday life easier for employees – internal chatbots

Public organisations can also create a chatbot to help employees by automatically answering their questions relating to human resources, IT, legal issues, etc. For example, the Val d’Oise Conseil Départemental implemented a HR chatbot in January 2020 to answer more than 4,000 employees’ most frequent questions and facilitate access to existing information and documents. This chatbot was created to replace a telephone call with an innovative and accessible tool. As such, Dimo, a digital assistant for Val d’Oise employees, provides generic and contextualised answers on topics such as leave and absences, social benefits, training, internal mobility, remote working, etc.

Then there is the city of Nîmes and its IT chatbot to help employees with Outlook, as they implemented a new internal messaging system. The chabot aims to make users more autonomous, but also to free up support services who were handling a large number of questions by email, telephone or in person. The city of Nîmes is now thinking of other ways their chatbot could develop employee autonomy in the future.

Chatbots are far more than a passing trend. They have gradually become an essential tool for departments that are overwhelmed by incoming requests – customer relations, human resources, IT, and more recently, public services who, in the midst of digital restructuring, are already reaping the benefits.

*Panorama of the digital transformation of the French public sector by Artimon, 2018

**Forrester, 2017

To fully understand the current crisis, we need to put it into context. This crisis is sudden; just 4 months after the virus emerged in China, 3 to 4 billion people are now in lockdown. We are facing an unprecedented situation. Almost every single country has been affected and forced to take action to fight the pandemic, making it difficult for countries to share their resources. This crisis is sudden and global, making it all the more challenging to overcome.

In this article, expert in business and crisis communication, Jean-Yves Rincé, analyses the Covid-19 crisis and shares his advice to help companies successfully communicate with their different audiences, in particular employees.

What is a crisis?

A crisis is a totally unexpected event that comes out of nowhere and plunges its surroundings into an extremely grave situation. It calls for urgent and highly coordinated action.

• unpredictable: it is impossible to predict a crisis’ start date, but it is possible to anticipate and prevent
• sudden: a crisis cannot be mitigated when it occurs, but we can react in an organised and calm
• manner
• urgent: a crisis calls for responsiveness, flexibility, and coordination

Crises are always awful but if the appropriate action is taken, they can be brought it to an end. Communication is key to achieving this.

What is crisis communication?

Crisis communication is not the same as crisis management. It refers to the various forms of communication employed to accompany a crisis as it develops and evolves and is addressed to the audiences impacted by this crisis, to guide them towards an end. Crisis communication does not
tackle the causes, but rather the effects, which are often psychological and sometimes physical.

Working on public opinion is a key part of this communication, i.e. reassuring people to avoid the risk of panic or doom-mongering, as seen in shops at the beginning of lockdown. It also consists in working on people’s behaviour, i.e. encouraging them to behave in a certain way, such as by
respecting lockdown, social distancing and barrier gestures.

Communicating during a crisis is complex because you need to address everyone, and quickly. Most businesses are currently faced with a particularly difficult situation as their two primary functions –
generating wealth and promoting social welfare – have been put on hold. The risk of losing their usefulness, legitimacy, and credibility is very real. It is therefore important they strengthen their roles and show that they convey positive values.

How should businesses communicate with their different audiences?

During these difficult times, companies have the opportunity and obligation to speak to their customers and employees, who may be struggling with a loss of bearings.

Understanding audience expectations

Employees are currently working from home, on reduced hours, unemployed, or exposed to the virus on a daily basis. In a totally unprecedented situation, 2 out of 3 workers are said to be working part
time or not at all.

They therefore make up a disoriented, and even anxious, audience for several reasons:

• the physical link between companies and employees no longer exists, as a result of lockdown
• daily employee work activities have changed, from their scope to processes. Managers must rely on
• an objectives and results system, in order to monitor employee progress
• the balance between work and family life can be difficult (shared living spaces, communication tools
• and time)
• management has drastically changed; employees may feel more pressure as a result of new
• formalities that did not exist before
• the vision employees have of the future is unclear, because they are focused on the present, a time of great uncertainty in terms of their health and professional future

As for customers, they are understanding but demanding. Understanding because the situation is difficult for everyone, so they can relate. But demanding because they expect businesses to offer solutions and compensation. They want to know what the conditions are for a return to “business as usual”.

There is a risk that companies could lose both of their audiences’ trust, possibly even leading to mistrust or suspicion. Their response to the crisis must therefore be very reassuring. It must build on what already exists and offer a solid perspective for the future.

Crisis communication: maintaining trust

Maintaining trust, at every level, in a company’s organisation, operations, management, and future, is an essential part of crisis communication. So, what 4 objectives should crisis communication work towards?

• Objective n°1: maintaining business continuity and integrity

Opportunities for employees and management to share and exchange should be facilitated and encouraged. These “online” moments help employees turn their focus back to the company, even if they are not physically present.

It is also important to answer employee questions, by implementing tools to collect and answer their queries.

• Objective n°2: facilitating work activities

Communication can be used to reassure, educate, and advise employees, so that they can complete their activities under the best possible conditions. The current situation is also an opportunity to look at how existing workstations can be made more ergonomic, and how processes can be improved.

• Objective n°3: establishing and embodying remote management

How can you be present for your employees when they need you, through communication? By inventing “telemanagement”, making it possible for managers to organise one-on-one chats, lead group discussions or brainstorming sessions. The tools already exist, but the frequency of such
exchanges needs to be defined.

• Objective n°4: defining and relaying a plan for the future

It is essential that companies define and communicate their plans for the future, after the crisis, whether for the collective future of the company or on an individual level. These plans should be authentic, realistic and presented with resolve. Do not make any promises you cannot keep, but also avoid being overly pessimistic, which would only serve to discourage your employees.

It is important to show that there is someone “at the helm”, with a real crisis exit strategy, and real perspective.

Crisis communication: being creative

The second key to successful crisis communication is creativity. Crisis communication cannot be based on yesterday’s issues or tools. You have to find other ways to communicate, to think out of the box. It is necessary to create new messages, to open yourself up to new horizons, based on existing foundations (the idea is not to destroy everything).

Overcoming a crisis means drawing on lessons from the past and inventing new things for the future. A company will find it easier to rally their employees if they convey a promise for change, for revival. Seize this period to establish a new kind of relationship with employees, and to ask them to contribute more. A crisis is an opportunity to build and exchange together on a number of issues.

The virus is a bit like a truth serum, administered to public authorities, companies and individuals, revealing their strengths and weaknesses. A crisis calls for renewed communication, because it is difficult to bring about change when things are going well, but possible in times of crisis.

If you want to know more about the benefits of a chatbot for crisis communication, click here.

Jean-Yves Rincé, expert in business and crisis communication.

35 years of experience in communication strategy consulting in-house, with consulting agencies or as an independent consultant. Founder of OPTIcommunication and lecturer at Sciences Po Lille and ENA.

The current situation is completely unprecedented. A large part of the population is under lockdown, forced to work remotely, temporarily laid off or on reduced hours. Another part of the population is out in the field, but in unusual working, transportation and environmental conditions. Business has been drastically reduced for some companies and administrations (for example the travel, leisure or transport industry), while others are struggling to cope with an increase in demand (retail, e-commerce, health and home-schooling services, etc.).

Staying in touch with customers and employees, informing them of the availability of services in real time – especially when things can change from hour to hour – is essential, whether to respond to changes in demand, or to maintain a vital link while waiting for a return to normal. The challenge is to ensure continuity of business, and constant communication with stakeholders (informing, reassuring and processing requests).

Implementing a chatbot to handle recurring incoming enquiries is particularly effective during a crisis. Read on to find out why.

A chatbot to best assist your customers

Whether deployed on your internet site, mobile app or instant messenger, conversational robots can quickly process simple customer questions and requests.

During a crisis, solicitations skyrocket and most customer services or call-centres are unable to absorb all the incoming requests. As a result, agents are overwhelmed, there are long queues, customers are dissatisfied and a company’s image can suffer (negative comments, tweets, Facebook messages, etc.).

Chatbots cannot answer all customer requests, but they are the best way to deal with a large number of incoming questions, for various reasons:

• they are available on all your communication channels 24/7
• they provide customers with instant and consistent answers 
• they are agile; companies can add new questions about the crisis in just a few clicks, and edit their answers as the situation evolves
• they relieve pressure on customer service teams, freeing up time for agents to handle more complex requests. You can avoid 80% of solicitations by covering 20% of the most frequently asked questions. A chatbot does not have to answer everything to be effective.

A chatbot to answer your employees’ questions

Internal chatbots, and HR bots in particular, have proven to be particularly useful during lockdown because forced remote working, especially in times of uncertainty, can raise many questions among employees.

HR chatbots, just like customer service chatbots, can instantly address your employees’ enquiries from their homes; remote working (hours, tools, obligations), temporary lay-offs or short-time working, wages and employee benefits (leave, lunch vouchers, retirement, transportation), childcare, sick leave, etc. The advantage of a bot is that questions and answers can be instantly updated as the situation evolves.

The General Data Protection Regulator came into effect on the 25^th May 2018. Any company domiciled in the European Union or processing personal data relating to EU nationals must comply with this regulation. Chatbots must therefore be compliant too.

But, in practical terms, what procedures should be implemented to ensure you’re your conversational robot complies with the GDPR?

Case n°1: The collected data does not allow to identify the user

If the chatbot does not collect personal data from the user, the aim is to dissuade the end-user from delivering any personal information during the dialogue that would allow their identification, by informing them of this before the dialogue begins.

Example with the OUI sncf chatbot :

Case n°2: The collected data allows to identify the user

This is the most frequent scenario, because to provide precise and personalised answers, the chatbot generally needs to access the user’s personal data. This is achieved via an automated process (webservice, SAML…) or non-automated (questions about the user’s identity during the dialogue).

Any information that makes the end user directly or indirectly identifiable falls under the category of so-called “data of a personal nature”, such as first and last name, address (for the delivery of a product for example) and e-mail address, as well as geolocation (for an itinerary), social security number (for health insurance), or even IP address (via cookies). Certain main principles of the GDPR must therefore be respected by chatbots:

• the obtention of user consent
• transparency about all processing operations performed
• the implementation of tools for the respect of users’ rights
• limit of data retention to that which is strictly necessary
• data security

Information and obtaining consent

Putting in place a welcome message informing users conversing with a chatbot for the first time of the collection and processing of their data is absolutely essential. This message should include information about why and how data is collected, the recipients of this data (chatbot publisher and third-party host if applicable), the retention period, as well as a reminder of the person’s rights. The message should be simple and easy to understand by all. It is recommended to add a link to your data protection policy (that can provide further details about data processing).

A Call To Action should feature below this message: “I accept” or “I understand”. Consent is given when the user clicks on this CTA. A variable should be associated in order to record the time and date of consent and keep a log of consent.

This is not mandatory, but we recommend creating one or several knowledge articles about the GDPR and data retention for your chatbot. This is an additional way to inform end-users of their rights.

Data processing and purposes

Personal data of end-users may be collected, recorded and stored by chatbot publishers for the companies that own the bot. They are collected primarily to meet two objectives:

• Personalise dialogues

Name, dialogue history, geolocation… Chatbots seek any information that will help contextualise the request, interpret it more precisely and provide a personalised answer.

• Improve the understanding of user requests

Machine learning enables chatbots to memorise dialogue situations to improve the matching between user questions and the knowledge base’s existing answers. The more analysed data, the more efficient the bot becomes. Under no circumstances should data be used for commercial purposes without informing users beforehand and obtaining their consent.

Hosting and security

The Datacentres used for hosting Cloud solutions should be located within the European Union, or a country deemed adequate by the European Union (see CNIL map). For “On Premise” solutions, a service security audit should be carried out (encrypted data, complex password, access permissions, intrusion tests, etc.)

Security should be reinforced for anything relating to minors and sensitive data such as health records or political opinions. Such cases entail additional obligations, particularly in terms of hosting (data stored with a host approved for this purpose).

Data anonymisation retention period

In order to comply with the GDPR, a retention period of a user’s dialogue history must be defined. At the end of that period, a purge or anonymisation of these dialogues should be carried out (we recommend a 15 to 30 day period at the most). A disclaimer should be put in place to inform users of the dialogue retention period.

Respecting user rights

Chatbot users, whether company customers or employees, should be able to access their data (right to portability) and to ask for this data to be removed (right to be forgotten). CTAs should be included in the chatbox for this purpose. When a user asks for access, they should receive an email with their dialogue history. When they exercise their right to be forgotten, all their dialogues with the bot should be deleted, followed by a confirmation email from the company informing the user.

Cookie management

The chatbox generates cookies in order to personalise the service (session cookies, login history). These cookies can be blocked by the end user at any moment, or by the company implementing the chatbot on their website, without hindering the instantaneous use of the bot. In order to comply with the regulation, it is recommended to manage chatbot cookies via a tag manager, in order to inform and obtain consent from users prior to integrating cookies. This is also true of any ancillary services generating cookies (for example, social media, videos, analytics, etc.).

Final advice: your data protection policy

In the ongoing interest of transparency, it is advisable to include references to your chatbot service in your data protection policy (or confidentiality policy) on your website, web application or intranet. This will provide additional and more detailed information to that included in the welcome message.

The GDPR entails a certain number of technical and legal constraints, but it is also the best way to gain your users’ trust, and to therefore encourage as many as possible to adopt your chatbot.

The 100 billion mark is in sight! Over the last 15 years, the French e-commerce market has not stopped growing. It reached 92.6 billion euros in 2018, a 13.4% increase compared to 2017, despite an unfavourable social context at the end of the year. The number of retail websites shot up by 12.6% in a year to reach nearly 200 000 active online stores. Yet despite the sheer number of players, 90% of the market lies in the hands of just 6% of e-merchants, approximately 10 000 websites (source: ZDNet).

To stand out, the ”small” players focus on innovation and the quality of their customer relationships. They are therefore increasingly turning to automation software to improve their business processes, make predictions, or improve their customer service. This article explores how AI and chatbots allow the e-commerce industry to reinvent itself in order to continue to grow and prepare for the years to come.

AI at the service of e-commerce

• Effective target marketing

AI is capable of analysing huge quantities of data. Used intelligently, they can complement marketing analytics and improve the brand’s target marketing. These solutions collect customer data (age, sex, location) and their behaviour (pages visited, time spent on website, number of clicks) in order to identify their consumer habits and promote the product(s) best adapted to the customer’s needs.

AI solutions can also help identify tomorrow’s trends. L’Oréal for example analyses thousands of images on social media in order to spot the products and details (colour, patterns) in vogue to add to their product ranges.

• Optimised logistics

AI may not have reached its full potential yet, but its scope of action within the supply chain is already considerable, whether in the optimisation of order preparations, transport, or stock forecasting. Take the giant Alibaba for example, who uses Cainiao, its logistics subsidiary, to pinpoint the most efficient delivery routes by using vast amounts of data. As such, they have reduced their use of vehicles by 10% and the distances travelled by 30%.

As for stock management, algorithms can be applied to identify early on which products are likely to sell faster or slower than expected.

• Visual searches

Truly revolutionary for customers – in just a few seconds, they can find the product they want with a photo taken on their smartphone. Fashion, home décor or furniture brands have been quick to adopt this new use, such as La Redoute who added visual recognition to its mobile app in 2017.

• Customer service and automated conversations

More and more e-commerce sites are equipping themselves with conversational assistants, otherwise known as chatbots. Featuring on their website, mobile app or messenger, these chatbots provide users with information before, during and after their purchase.

The rest of this article will focus on the above, and how chatbots can help improve an online store’s customer service.

Conversational AI for better customer service

Conversational Artificial Intelligence can be defined as a machine (a computer programme), with the ability to hold a conversation with a human at different levels of complexity. Chatbots are one of these conversational AI applications.

• Chatbots, from virtual assistants to personal shoppers

“The more you tell, the more you sell.” The rise of chatbots and their multiple uses in the e-commerce sector seems to confirm this adage.

Over the last 10 years, chabots have made their way onto most companies’ websites, mobile apps and messaging platforms. They appear in the form of a pop-up window and usually begin by defining their scope (chatbots are not supposed to answer all questions): product recommendations, after-sales queries, logistics questions (cost, delivery times, refund policy, etc.). And when the chatbot does not know the answer, it hands over to a human operator.

Today, chatbots are far more popular and reliable than you may think. According to a recent Ubisend report, 1 in 5 customers consider buying a product or service via a chatbot.

We will now take a look at how chatbots can improve the customer’s e-commerce purchasing experience.

• Availability and instantaneity

Availability is a chatbot’s best asset! They provide answers in real time, 24/7, a considerable advantage bearing in mind that consumers tend to make their online purchases in the evening, with 19% of sales between 6 and 8pm (source: IFOP).

A question about delivery, a return or refund? A chatbot’s availability and instantaneity is what appeals most to users. 71% of consumers say they would use messaging apps for customer support (Hubspot, 2017), primarily to ensure the swift resolution of their problem. The quality of customer service is one of the main reasons a customer stays, returns or decides to leave an e-commerce site.

In addition to their availability, every chatbot is endowed with its own personality, designed in advance by the brand implementing it (tone of voice, expressions). Chatbots are therefore also their brand’s spokesperson and can inspire user sympathy. Bots are never in a bad mood and can sometimes even make you smile.

• Personalisation et product recommendation

Beyond their ability to answer user questions about products, terms of purchase or logistics, chatbots can create a range of content that is adapted to each visitor.

Sephora’s messenger chatbot is a perfect example of this. Simply answer a few quick questions and gain access to personalised beauty advice. Providing appropriate and relevant content both satisfies existing customers and converts those who are on the fence. Visitors to a website feel that they are being assisted, thus building a relationship of trust. Chatbots engage the user and encourage them to buy. They can also promote other products that may interest the customer, otherwise known as cross-selling: “if you like… then you will like…”. They can also upsell by offering personalised promotions for each visitor in order to encourage them to make another purchase.

As such, a well-designed chatbot can improve an e-commerce site’s conversion rate, whilst decreasing their bounce rate, one of the biggest challenges faced by online stores.

By asking the right questions, a chatbot can recommend personalised products based on the user’s answers or saved preferences.

LEGO’s chatbot messenger Ralph is a perfect example of this. Ralph asks a series of questions about the person’s age, budget, or favourite topics in order to find the perfect gift.

• Smooth customer service and cost reductions

Chatbots are useful for buyers and future buyers, but they also strongly benefit the companies implementing them; increase in productivity, cost reductions, and employees who feel valued.

Customer advisors are no longer required to answer the simplest and most frequently asked questions; the bot does it instead. In many cases, a significant drop in the number of inbound contacts can be observed following the implementation of a bot. Lower contact rates mean lower costs. The cost per customer request for an agent is said to average around 6€ in B2C and over 11€ in B2B (Harvard Business Review, 2017).

Automating answers that were once provided by agents does mean that the company wants to replace them. Covering 20% of the most frequently asked questions avoids 80% of enquiries. Chatbots do not therefore need to answer everything to be efficient. In certain cases, a hundred or so knowledge articles are more than enough. The agent works the chatbot and takes over for the more complex queries.

When a brand implements a chatbot, they ensure both their conversational strategy with customers, and their conversion strategy. A well-treated customer, whose expectations are met, is a loyal customer. Research, data analysis, classification, advice… Humans have endowed bots with their intelligence. And this is only the beginning of what bots will be able to do tomorrow.

More and more companies are implementing conversational robots to make their employees’ daily lives easier. These robots primarily answer questions about human resources, but not only! Bots focusing on office automation or legal support are also on the rise.

They provide employees with general answers about the company, or personalised information relating to their individual status. If an answer requires further explanation, the robot can refer employees to electronic documents, explanatory guides or even a specific intranet page. Actions such as requesting leave can also be carried out via the bot, since it is connected to the company’s IS.

Chatbots can be deployed on all of a company’s internal communication channels; intranet, SharePoint space, mobile apps or messaging platforms such as Slack, Skype, Messenger, etc. No longer exclusively reserved for large companies, chatbots are attracting more and more small to medium businesses too.

Why implement an internal bot?

In order to improve employee efficiency, a company may decide to set up a chatbot to answer their most frequently asked questions 24/7. This also allows to optimise support functions’ costs since the chatbot relieves them of the repetitive aspects of their work. Covering 20% of the most frequently asked questions helps to avoid 80% of requests. A chatbot does not therefore need to answer everything in order to be efficient; in some cases, a hundred knowledge articles is enough.

Chatbots for internal use present a number of advantages for both the employee and company implementing them:

• Increase in employee productivity thanks to immediate access to information at any time of day
• Relief for support functions resulting in improved efficiency on important issues
• Empowered employees performing tasks with added value, thereby reducing turnover (by up to 28% according to the HR practical guide ParlonsRH)
• A positive and innovative company image, attracting new talent and building employee loyalty

The rest of this article will focus on the most frequent chatbot use cases.

A bot for Human Resources

Recruitment

Recruitment chatbots simplify a HR department’s work by performing an initial automated filter on the hundreds or even thousands of applications received. Analysing each one, they can determine a candidate’s level of compatibility with the company and position to be filled. Chatbots are more objective than humans in their analysis; they only take skills and experience into account, and not the criteria considered discriminatory. Randy, the recruitment group Randstad’s virtual assistant, is the perfect example of a successful recruitment chatbot, voted chatbot of the year at the “Grand Prix du Social Média” in 2018.

Chatbots can also perform personality, skill or motivation tests by interacting with applicants. Others help the selected applicants to prepare for their interviews, such as the BNP Paribas chatbot “Dr Job”.

Onboarding

When starting a new job, many employees do not dare ask questions. Chatbots facilitate onboarding by indexing all the information that employee may require; administrative assistance, practical information, rules and regulations, company policy, etc. The bot can even send notifications to remind new employees to fill in a form or complete a task.

Administrative management

A HR bot’s role is to make employee lives easier by answering their most frequently asked questions and automating simple actions. Employees can obtain information about their rights (remote working, internal mobility, company agreements, training courses, etc.), as well as personal information linked to their own status within the company. Connected to the company’s HRIS, the bot can retrieve information from various systems and provide each employee with their leave balance or last payslip for example.

Lots of tasks can be performed via a chatbot; request leave, report sick leave, amend personal information, request training, issue an expenses report… A single entry point for users and a considerable time saving tool for HR!

Practical services

To make your employees’ lives easier, there are also “concierge” chatbots, providing a wealth of practical information; how to reserve a parking space or hire a vehicle, when to drop washing off at the drycleaners, company childcare opening hours, etc.

According to a study carried out by Olfeo in 2016, French employees spend one hour per day dealing with personal issues at work. By handling personal requests such as booking a taxi or meal deliveries, concierge chatbots can truly assist employees. As well as saving time, and therefore efficiency, chatbots are an excellent way to develop a company’s employer brand.

Change management

Implementing a conversational assistant can prove to be particularly useful when a company is undergoing any major changes, such as the implementation of a new agreement, the deployment of regulation such as GDPR or the introduction of tax deducted at source. This helps to smooth out any peaks in requests and avoid feelings of frustration amongst employees.

Total for example implemented a new chatbot, Hector, to support employees during the implementation of a new seniority agreement. The bot provides information about the agreement’s conditions and employment records.  It can also calculate the employee’s eligibility depending on their age and quarterly contributions.

HR bots meet a number of different needs. As such, their use cases can be varied. If you would like to know more about Human Resource Managers’ perceptions of bots, look at our HR Chatbot Observatory survey.

A bot for IT

IT services and helpdesks are often overloaded and struggle to respond quickly to all IT support requests. Long waiting times and overly technical explanations can sometimes be a source of tension amongst departments. Implementing an IT chatbot can significantly reduce the number of incoming tickets and improve the request management process. The bot answers first and second level questions related to IT and desktop applications; password management, software assistance, hardware orders, account management, etc. And when human intervention is necessary, you can directly create and follow a ticket via the bot!

IT chatbots are even more useful in the sense that IT requests are often urgent, leaving employees stuck and unable to continue working. In addition to immediate responses, the 24/7 assistance provided by the tool is a genuine asset, particularly for people working staggered hours. The virtual assistant makes users more autonomous with regards to simple tasks, by capitalising on existing company documentation; product tutorials, problem solving guides, step by step instructions, etc.

Implementing a chatbot covering IT and helpdesks considerably reduces the number of tickets created, sometimes by up to -25%. Just like a HR bot, an IT bot is very useful in the event of major changes within a company such as an IT system migration. A lot of companies have supported their employees during a migration to Office 365 by implementing a dedicated chatbot, for example.

A bot for the legal department

“Legalbots”, the latest internal chatbots to appear, are on the rise, relieving legal departments of time consuming tasks. The chatbot answers the most frequent legal questions and provides practitioners with the documents in force to carry out their work, such as the framework for a confidentiality contract. This applies to a variety of areas; social law, criminal law, commercial law, contract law, VAT, corporate tax, GDPR, intellectual property, competition law, etc.

The legal chatbot implemented by EDF in 2017 is a great example. It covers all areas of corporate law and generates customised contractual frameworks, confidentiality agreements adapted to different situations and transactional protocols. As well as freeing up time for legal practitioners, the bot allows to homogenise answers and make them easy to understand for all employees. EDF won an award for legal innovation,  le prix de l’innovation des directions juridiques in 2018 with this chatbot.

Conversational robots can also address other use cases. The Société Générale’s chatbot deals with finance, communication and compliance for example. Bots providing technical assistance are also widespread in building or energy companies, such as EDF and Engie who have implemented internal chatbots relating to the photovoltaic field. To go one step further, please feel free to download the ADP Group interview about their implementation of IT, HR and legal chatbots.

You may already have heard of callbots, new conversational robots that can be reached via telephone. Recently deployed in companies facing a high volume of calls (insurance, telephony, transport, e-commerce, hospitality…), they automatically answer customers or future customers’ questions and requests. They also qualify these requests and can transfer them to skilled advisers, as per the operational strategy in place.

These new bots replace Interactive Vocal Response (IVR) systems, where the user presses buttons on their keypad to reach the appropriate service: “To speak to a sales representative, please press 1. To reach our customer service, please press 2. To be connected to an agent, please press 3…”

Unlike IVRs, callbots dialogue with users in natural language. Read on to find out how.

How do they work?

Like chatbots, callbots function with a Natural Language Processing (NLP) algorithm. This algorithm makes it possible to render the conversation as natural as possible, thanks to the management of digressions, “small talk” and personalised responses, according to context and user history.

For a chatbot to become a callbot, two extra steps need to be added:

• Voice recognition (Speech-To-Text), the input which allows to capture the request and analyse it
• Voice synthesis (Text-To-Speech), the output which allows to deliver the requested information vocally to the user

Callbots are also connected to the company’s various web services (IS, CRM and ticketing platforms), to personalise the responses and trigger actions such as escalation to an adviser or sending a confirmation email/text message.

Extra work is required in comparison to setting up a chatbot; voice design! A voice-user interface is very different to a text one, and the bot needs to be adapted accordingly. Tone of voice, nuances, interruptions… In order for the conversation to be as natural as possible, the callbot must adapt to the user.

To find out more, read our article about conversational design.

Why should you implement a callbot?

By implementing a telephone conversational robot, you can meet the customer, adviser and company’s needs all at once:

And what about humans?

Like any other chatbot, callbots are not destined to replace advisers, but to work alongside them. The human role is essential, the bot acts as a complement to their work and therefore allows them to focus on more important tasks.

At the moment, whether textual or vocal, bots are unable to manage all the tasks performed by man. As soon as the conversation becomes more complex, the presence of an agent is essential to ensure a good customer experience.

And even if technologies are evolving rapidly and use cases multiplying, it is not in a company’s interest to let robots manage their entire customer relationship.

MACSF case study

MASCF, an insurance company for health professionals, implemented a callbot in December 2018 in order to extend the opening hours of its customer service and pre-qualify incoming requests. The aim? To improve the experience of (future) policy holders, increase sales and make the processing of requests smoother.

Thanks to the callbot, named Marion, if a user reports a claim in the evening or at night, they are automatically called back the next day to confirm, saving a considerable amount of time for both the insurer and policy holder. Users can also take out new car insurance via the callbot, effective immediately. At the end of the conversation, the bot sends the user confirmation by text message and integrates their request into the IS so that it can be processed by the correct management service.

During the very first month following the launch, Marion answered 2000 calls with an 89% intent recognition rate. The next steps will be to optimise the existing use cases, and to add new use cases to implement a permanent robot (working and non-working hours).

Last April, MACSF won the “Argus d’or” for its callbot, a prize rewarding the best innovation initiatives in the insurance industry.

If you would like to know more about our dydu callbots, please don’t hesitate to ask for a demo.

Automate repetitive tasks

At what salary would you be willing to perform the same repetitive tasks? In “Les enjeux psychiques du travail” (mental issues of work), Pascal Molinier describes the work of telephone operators as a pathogenic source of neuroses and anxiety.

Indeed, people working in call centers are, most of the time, answering the same questions. This is less rewarding by definition than those involving research work, stimulating our awakening, curiosity and creativity. As a result, the need to automate the redundant is important. The chatbots come into play! When, how and at what cost?

A solution? Chatbots!

Chatbots, conversational agents, or virtual assistants, have been enjoying a massive expansion around 2010 with Facebook and Microsoft. After a few mixed or bad experiments, the most famous of which will surely remain Tay, object-oriented programming is today effectively crystallizing concepts such as “conversation”, “digressions” or “intentions”, both written and oral. Thanks to the Natural Language Understanding, artificial intelligence allows a more fluid, multi-supported interaction than ever before, and above all, allows to meet the initial need: to give the right answer quickly.

Goal: optimizing time management

Whether we are looking for « a ticket for 2 people from Paris to Toulouse for Valentine’s Day weekend », or requesting « paid leave next Monday and Tuesday », conversational agents offer considerable time savings to users. In the example above, it takes an average of 1 minute to complete your trip on a ticket booking website. Formulated as such, it takes 15 seconds to type!…and only 5 seconds to speak! In companies, mastering internal jargon is a decisive step in ensuring good communication. Rather than interrupting our manager daily on such questions, why not ask them to the new chatbot?

Towards a diversification of uses

Luxury, finance, transport, energy, industry… medical? To meet their goals while complying with the rules of the targeted sectors, chatbots interface with facebook, skype, hangouts, but also slack, whatsapp… and are slowly invading show rooms to living rooms through cars on-board computers (Renault, Nissan, Über). In after-sales service, chatbots became the first factor for improving customer relations in 2017, with 80% of right answers (source: PwC). The CAC40 companies have understood this and are currently all equipped with at least one chatbot. Thus, as long as a list of the most frequently asked questions on a given topic can be compiled, the best solution will always be the cheapest one that works.

Free and GDPR-compatible self-service solutions exist, so why not using them?

What exactly is a voicebot?

If you missed the media frenzy surrounding voicebots, here’s a quick catch-up; a voicebot is a vocal conversational robot. They are equipped with the same capabilities as chatbots, meaning an understanding of natural language and the ability to respond to users automatically. The only difference lies in the interaction interface, which in this case is vocal. The most famous voicebots include Google Assistant, Alexa, or even Siri.

Some of these vocal assistants can be found in recent car models, smartphones or even some connected speakers such as Google Home, Amazon Echo or HomePod. In terms of innovation, they are the very latest, must-have technologies, that companies dream of acquiring in order to offer a unique and out of the ordinary user experience.

Voicebots, simple gadgets or future technological revolution?

Despite an increase in vocal applications and their different uses, user satisfaction is not always guaranteed. As such, many of us still shy away from using these conversational robots in our everyday lives, primarily for the following reasons:

• They require behavioural change; not everyone is comfortable talking to an object. Just look at the number of people who use the vocal assistant on their telephone (do you?)
• We don’t know what to ask. Or how to go about it. What can they understand? The experience is disappointing when the voicebot doesn’t know how to respond (even more so than with a chatbot!)
• We can’t have a “real” conversation with them. They respond to simple vocal commands such as “Switch the light on”, “Turn the music on”, or common questions such as “What’s the weather like?”, but you cannot enter into an actual dialogue with them
• We don’t always understand their utility
• We don’t trust them (am I being listened to? what happens to all my data?)

So, with this in mind, why are voicebots expected to be a revolution in years to come?

Voicebots, tomorrow’s must-have technology

> They free up your hands

Voicebots allow users to multitask, enabling them to complete an action without having access to the control interface in question. Vocal command is particularly useful in cars, because it allows the user to drive whilst using their smartphone or changing their vehicle configuration.

They are well suited to other everyday situations; who has never covered their smartphone with flour in the kitchen whilst trying to read the next part of a recipe? When it would be so much easier to say out loud, “What’s the next step?”.

> They facilitate accessibility

Vocal assistance makes technology accessible to everyone; in particular young children, old people, people with disabilities, or even illiterate people. By simply using their voice, a multitude of actions and information become readily available to them.

> They personalise

Voice is a strong marker of a person or brand’s identity. For a company, it is a factor of differentiation and personification. Just as you would choose your logo, slogan or visual identity, you choose a voice for your brand and Voicebot. It should be easily identifiable by its tone, character and personality. Who in France wouldn’t immediately recognise Simone, the SNCF train service’s voice? Voice is becoming a major topic of consideration for communication and marketing teams.

> They ‘re-enchant’ the user experience

Users are more and more demanding; they want novelty, new experiences, sensations and emotion. Which is precisely what voicebots have to offer. If the user experience is thought out and built with the right users in mind, then the final result will be very impressive. This is the case for insurance company MACSF’s telephone voicebot, for example. Deployed for their beneficiaries, it won the Argus prize that rewards innovation initiatives in the insurance industry.

The stakes are just as high as the risk of a disappointing experience. But when the voice design concepts are respected, and the instances of use and processes well built, then voicebots are the easiest and quickest ways for users to access your services.

A few Voice Design best practices

Creating a good voicebot is not trivial. A design phase specifically for the use of voice is required, known as Voice Design, in order to produce a Voice User Interface (VUI) worthy of that name. To guarantee a unique vocal experience, certain principles must be respected:

• Guide users interacting with the voicebot, on what they can and cannot do, but also when the bot does not understand something. “Sorry, I can only help you in the kitchen” will always be better received than an abrupt “I don’t understand”
• Take the context of the conversation into account (user profile, time, location, voicebot conditions of use, etc.), in order to enrich the dialogue with relevant suggestions
• Be brief and concise; a user’s time is precious. Avoid long monologues and leave pauses for the users to express themselves too
• Give the voicebot a personality to increase its emotional impact (and therefore the brand’s)
• Make the conversation feel human; tone of voice, nuance, interruptions, human expressions (“hmm, yes I see”, ”perfect”, “noted”)
• Divide up the construction of the voicebot to improve the VUI progressively
• Favour sentences and questions leading to precise answers; “Please specify if you would like to book a hotel or a taxi?” would work better than “Would you like to book a hotel or a taxi?”

The list is long and non-exhaustive. Each voicebot is unique, and therefore needs to be adapted, whenever possible, to each instance of use.

So, will Voicebots replace Chatbots?

Speaking is 3 times faster than typing on a keyboard. Speech is mastered well before writing and is our most natural form of communication. Since a UX without visuals frees the user from interfaces, many believe that this will become our new standard.

That is not to say that Chatbots have signed their death warrant. On the contrary, their use can be complementary to Voicebots, depending on user preferences and the conditions of use. For example, a chatbot will be favoured in noisy or public places for discretion and confidentiality reasons. They still have a bright future ahead!

For a long time, we have enjoyed imagining a world in which robots and artificial intelligence come to replace man, from a functional, intellectual and even emotional standpoint. And in the worst case scenario, take control and try (in vain) to eradicate the human race #terminator. Thankfully, this is not the case.

When AI resurfaced as a major topic of public debate in 2015, a legitimate question emerged; are our jobs under threat? Will AI eventually replace some roles ? An (unsurprisingly) divisive debate that has led some to speak of a job market transformation. So, what’s the reality?

An initial observation

One thing is certain, numerous technologies are in vogue at the moment (machine learning, automated data collection and analysis, personal digital assistants, autonomous vehicles…), but some form of human assistance is always required to fully benefit from their functionalities. The technology alone is not mature enough to be used as it is, and research and development is always needed to optimise it. Within this context, new companies taking advantage of the AI boom are emerging, thus creating specific jobs to carry out their missions.

New jobs are also being created amongst the bigger players (digital innovation hubs, AI…), thus offering new opportunities. And those who have already put in place an AI-enriched solution (such as chatbots), are seizing the opportunity to free their teams of repetitive tasks, whilst also saving time and money for the more strategic ones.

As you will have understood, AI allows companies to increase the number of employees in order to optimise their performance. And not to lay them off.

And from a sociological standpoint?

As everyone knows, change is scary, mainly because we don’t have control (over everything). This change is even scarier because we are not replacing mechanical functions with machines, but intellectual functions with AI.

The major movements that changed labour organisation immediately come to mind (Fordism, Toyotism…). These principles were also created to optimise company performance. Take Fordism for example; new processes led to a boost in productivity and a decrease in costs. And the icing on the cake? Wages went up to increase consumption. And yes, a loss of worker skills was also observed, forcing employees to retrain.

And in general, positions nowadays in which people acquire a life-long skill are increasingly rare. A lot of people will (or already have), at some point in life, need to retrain. Which is what we will see more and more with AI; jobs changing, others being created. For students, new specialisms have also emerged, following this trend. “Nothing is lost, nothing is created, everything is transformed” (or almost).

Training to save jobs

Within this context, training now plays, more than ever, a decisive role. It enables employees to develop their expertise or, if necessary, to be reassigned elsewhere, depending on the company’s needs. Adaptability is set to become a considerable asset on the job market. The ability to learn something new mid-career and change jobs quickly will be crucial. The companies who understand this will not only avoid redundancy waves, catastrophic for their image, but also retain staff. And all the while, develop their employer brand, a virtuous circle for attracting new talent. So, to all companies, big or small, may the training begin!

The rise of chatbots and other conversational robots is nothing new. In 2015, companies began deploying them massively on their websites or mobile apps, in order to relieve pressure on their call centres.

Since then, their uses have multiplied; product recommendation, employee support (IT, HR, helpdesk, legal), welcoming visitors, home automation… Today, numerous companies are equipped with one or more bots, and are always exploring new ways to use them.

Far more than just a passing trend, these virtual assistants have real advantages for both companies and end users; rapidly accessible information, reduced wait times, a smoother processing of requests, reduced contact volume… Yet, sometimes users are disappointed; misunderstandings, poor suggestions, repetitions… Some bots clearly don’t live up to our expectations.

To avoid user frustration, it is essential to work on your bot’s conversational design. Read on to find out how.

Spending time on conversational design to encourage the adoption of your bot

Conversational design includes all the steps necessary to making interactions between a human and a machine as smooth and efficient as possible. To create your User Interface (UI), you must first define the User Experience (UX).

UX takes all of your end users’ needs into consideration, to offer them the best possible experience. Your bot should adapt to them, and not the other way round. UI is the experience’s visible medium, a direct link between the user and product. A good UX/UI makes all the difference.

Understanding your users and their needs

The first step when creating a chatbot is to put yourself in your end users’ shoes by creating personas. In other words, identifying your target and establishing their personality, habits, motivations and frustrations. Next define all the possible expectations from these personas. Only after this step can you start to create a bot’s knowledge base.

Define your chatbot’s personality

Name, avatar, conversational style, tone… Everything should be thought out with your end users’ profile in mind, sticking as closely as possible to your company’s brand image. As such, a chatbot can be funny, friendly or more formal.

Knowing the context

Your bot should be able to contextualise user intent and bounce back with suggestions that make sense according to the user’s profile and the bot’s use. You can also inject your bot with “emotions”, via an animated avatar or emojis. This will make it appear more human and friendly, and users will enjoy chatting with it!

Guiding your users

When starting a conversation, your bot should first and foremost introduce itself and then specify the perimeter within which it operates, to avoid any frustrations. The bot is not supposed to have an answer for everything! The conversation could start for example by, “Hello, I’m Leo, your virtual adviser. I’m here to answer any questions you may have about tax deducted at source”.

The chatbot should also guide your users during a dialogue, through the misunderstandings and potential actions and requests. Should an ambiguous situation arise, the bot should be able to specify the user’s intent and transfer them to a human when unable to answer. All of these different scenarios are defined right at the beginning of the project and integrated into the bot’s back-office to be able to handle 80% of end user interactions.

Generating smooth conversations

To guarantee the best possible experience, your bot must be able to generate a smooth conversation, as natural as possible. Slot filling can be an interesting functionality to use here; it enables the bot to gather information or elements throughout the conversation, without necessarily following a question/answer sequence, that are needed to provide a response to a user’s intent.

For example, to respond to the intent “I would like to order a pizza”, you can configure a “toppings” slot and a “number of people” slot. If the user says:

• “I would like to order a pizza”: the bot asks the user to specify which toppings and for how many people
• “I would like to order a cheese pizza”: the algorithm fills in the “toppings” slot with “cheese” and the bot asks the user to specify for how many people
• “I would like to order a cheese pizza for 4 people” OR “I would like to order a pizza for 4 people with cheese”: the algorithm fills in the “toppings” slot with “cheese”, the “number of people” slot with 4, and the bot can continue with the order

And what about voice?

You have probably already exchanged with a vocal assistant, which is hardly surprising given that they are a hot topic of discussion at the moment and have invaded our everyday lives; on our smartphones with Siri or Cortana, via connected speakers like Google Home and Amazon Echo, or other connected devices. Gartner estimates that 20% of interactions will occur via vocal assistants in 2019. A figure that makes sense when you consider that talking is 3 times quicker than typing on a keyboard.

Voicebots are based on the same technology as NLP (Natural Language Processing) and endowed with the same capacities as chatbots, the only difference being that their interactions are vocal. They do however also include two extra steps: voice recognition (Speech-To-Text) and vocal synthesis (Text-To-Speech).

Voice design, an essential step

To create a voicebot or a callbot (a voicebot via telephone), you can follow the same steps as for a chatbot.

Interactions between humans and robots are nevertheless different because the user interacts using speech instead of writing. The choice of words, syntax, utterances and expressions are thus all the more important. In this step, it is essential to understand the difference between talking and writing before creating your bot’s knowledge articles and the related decision trees.

The keys to a good Voice UI, according to dydu:

• Choosing the right voice
• Being as natural as possible: tone, nuances, interruptions
• Keeping it short and concise: the user’s time is precious
• Being cooperative and empathetic
• Making suggestions
• Varying words and expressions to avoid repetitive sentences
• Using and understanding common words and human expressions: “Hmm, right, I see…”, “Perfect”, “Noted, thanks”
• Testing, retesting and re-retesting the bot ?

Whether used alone, or in addition to text and images, speech can truly offer a rich user experience, as long as you respect the best practices when it comes to voice design.

First off, don’t panic. You are neither the first, nor the last, to whom this has happened. Lots of chatbots have fallen into disuse, removed just as quickly as they were created, feeding into a list of long-lost chatbots that gets bigger by the day. Cast your mind back to Anna from Ikea, Emma from NotrePharma, Hugo from Mycoachfoot, Eva from Free, Jiminy from Gites de France or even Maxime from Propriétés Privées, to mention but a few… None of them survived, all for different reasons. In this article, we will try and explain the ins and outs of their sudden disappearance.

You want a chatbot “because it’s trendy”

It’s true, chatbots are all the rage. Just like a pair of sneakers, you want your very own chatbot, to show the world your innovative style, that you’re an “early adopter” or even a “techno lover”.  Or maybe you simply want to be like your competitor who has a really great (and popular!) chatbot. But beyond all the hype, you don’t really know what you want your chatbot to be used for, or what you can make it do. A chatbot should serve a specific purpose – to help visitors navigate your website and advise them on their product selection, generate certificates for your associates, or even incite customer loyalty with good quality content. This purpose is your compass – it guides you in your strategic decisions during the project, but also in your design choice and your bot’s personality. And, above all else, it helps you to define the perimeter within which your chatbot will operate.

You did not correctly define the perimeter within which your chatbot operates

A chatbot cannot (as of yet) answer every single question we ask it. It is therefore important to define the scope of questions you expect it to address. These question and answer duos, known as “intents”, are listed ahead of the project, and stem from the objective assigned to the bot at the very beginning, hence the importance of carefully defining its perimeter. Your current knowledge bases, such as your FAQs, can serve as a starting point for this list. If you have not yet capitalised on your knowledge, you can get started by setting up a live chat – operators who reply to user questions – in order to identify the most frequently asked questions and then automate them via the chatbot.

Whatever your strategy for building up your list of intents, we recommend testing it first with a pilot phase. Place your chatbot in the hands of a small group of users who are already familiar and onboard with your project, then explain that they are helping you to make improvements. The more questions they ask, the more the chatbot will learn, and the less risks you will take when opening it up to all your users. A 3 month trial period is sufficient to cover 95% of the questions that your bot will be asked later.

You forgot to explain your chatbot’s role

This is often the main pitfall, the fatal error that is pretty much guaranteed to lead your chatbot awry. As mentioned above, a chatbot should serve a specific purpose. Your users cannot guess what that is if you don’t explain it! You subsequently have a 50% chance that a user’s first 3 questions will be off-topic. And yet that is precisely the moment when a user adopts a chatbot, or not, as a new digital tool. 3 wrong answers and you can say goodbye to your user, he or she won’t be coming back.

“Hi, I’m Tom, your new chatbot. I’m here to answer your questions about tax deducted at source.” “Hi, I’m Theo, an apprentice robot, and I might be able to help you. Do you have a question about transferring your phone number? Then ask me: “how do I transfer my number?”, or “what’s a PAC code?”. “Hi! I’m Leo, your virtual advisor! I’m here to help you with your subscription. ? If I’m unsure about anything, I’ll pass you over to an adviser if there’s one available or explain how to get in touch.” These introductory phrases are reassuring for your users and guide them as they interact with the bot for the first time. They allow you to avoid misunderstandings and people getting frustrated, stressed and eventually giving up. This is especially important given that a failed conversation can generate negative feedback on social media, or even negative buzz. Don’t let a poorly configured bot damage your brand image!

After the launch, you thought everything was (finally) over

You’ve identified your bot’s objective, the perimeter within which it operates, your introductory sentences are configured, the pilot phase with your trial users is over… Phew! You can finally get some rest. Think again! The adventure has only just begun… ?

Your bot is now exposed to outside aggressions and provocations from your end users, who will make sure to put it through the mill. It is therefore necessary to monitor interactions in order to analyse the unsuccessful conversations, identify what’s falling through the net, and take a close look at the user satisfaction ratings. Reasons for dissatisfaction include missing, poorly written or lengthy answers, all of which can irritate. Reasons for satisfaction include direct, to the point answers and a bot close to its users. The constant improvement of your chatbot is essential for post-launch survival and to ensure that users regularly come back.

We hope to have shed some light on the pitfalls that could undermine your chatbot launch and scare your users off. At dydu, we support our clients every step of the way to help them get their chatbot up and running, and we can guarantee a successful launch. The proof – 7 years later, our first chatbots are still very much present and answering thousands of questions every day, much to the end user’s delight. If you would like to know more or book a demo, don’t hesitate to contact us!

This is the distant memory that comes to mind when we read, through slightly clenched teeth, the wild imaginings surrounding the wonderful industry we devote our time to here at Do You Dream Up, otherwise known as “chatbots”. As though there were no particular resemblances between a chatbot and its master!

Amongst those commenting on the arrival of chatbots, 3 schools of thought can actually be observed:

• The first wants to know what a “chatbot” is. This is one of the easier ones to answer. A chatbot is a “conversational agent”, a contraction of “chat” (as in a conversation) and the “bot” from “robot”. Our chatbots are particularly useful for maintaining relationships with clients, employees, partners… basically anyone who visits your website or app.
• The second denounce the supposed link between chatbots and Artificial Intelligence (AI) and criticises left, right and centre the “idiotic” nature of chatbots. chatbots are in constant demand since, according to Gartner, a customer will manage 85% of the relationship with a business without interacting with humans by 2020. This school of thought therefore aims to dispel those looking to take advantage of the buzz. And indeed, there’s quite a buzz around chatbots. Everyone seems to want one.  We won’t slip into systematic criticism. Chatbots resemble those who create them.
• The third school of thought sees chatbots as the Horsemen of the Apocalypse, coming after our economy, jobs and even the human race itself… There are a lot of doomsayers of this kind amongst consultants. Which is normal, identifying a problem is part of a consultant’s job. Fixing it, is ours.

So, how about we take things down a notch?

Chatbots are very clever and are becoming even more so, but they are also our creations. Since Mary Shelley and Frankenstein, a whole body of literature has devoted itself to convincing us that man creates and gives life to ideas that will ultimately destroy him. In our daily lives as modest and enduring innovators, our one hundred or so clients in France and elsewhere, work hard to give their customers, partners or employees a chatbot able to perform the services the user expects of them.

A few examples:

• The “coach” chatbot can help an employee in their first steps when thinking about career mobility. The “coach” chatbot is not a coach, but it helps the user to ask themselves the right questions.
• The “legal” chatbot can help teams to finalise or analyse the clauses of a contract. The “legal” chatbot is not a lawyer, but it can unload a great amount of excess work in the form of frequently asked questions, thus allowing legal departments to concentrate on projects to which they bring added value.
• The “IT” chatbot is the one you turn to when you forget your password to turn on your computer. The advantage of a chatbot, is that it’s available day and night. It will even interact with you more and more if it can ascertain whether you are panicking, angry or just in a spot of bother. And if your problem was to be more substantial, it would hand you over to a competent computer specialist, along with their specific skillsets, rather than face the dispositions of distracted or even dishonest users all the livelong day.
• The “HR” chatbot is not a Human Resources manager, but it will be able to tell you how to apply for leave, and even do this for you, without you needing to know the procedure by heart. In short, a chatbot can be a diligent and efficient assistant.

All these examples illustrate some of the good reasons organisations, whether public or private, and of course brands, create chatbots. We live in a world of increasingly intense, fast and borderless services and interactions. Chatbots can be resistant, quick and without borders, as long as the organisations or brands creating them carefully consider their content strategy, right from their implementation. Afterall, there’s no salvation without content!

Recent months on the web have seen an increased passion for subjects relating to machine learning, automatic learning algorithms or, in broader terms, artificial intelligence.

These technologies have been around for a long time, but in just a few years have expanded rapidly, becoming more and more sophisticated and offering increasingly high levels of performance.

Many players in the digital industry have chosen to invest in research dedicated to these technologies, including giants such as Google or Facebook. The research agency TechSci Research expects to see growth of 75% in the Artificial Intelligence sector for the period 2016-2021.

AI is now used in many domains, and in particular in the management of automated interactions in customer relations contexts: these are known as chatbots, or conversational robots. According to TechSci Research, a third of Americans have already used this technology in the form of this new kind of personal assistant.

Data the priority

Various experiments have led to the emergence of a certain number of myths around artificial intelligence, and in particular its self-learning abilities, some of which generate fear. We can of course refer to the victory for the AlphaGo machine against Lee Sedol, champion of the game Go, leading to fantasies of machines being able to surpass humans.

Yet machines are programmed by humans and are dependent on humans in order to work effectively. Another example: last spring, Microsoft performed a machine learning experiment with a chatbot which taught itself based on its interactions with Twitter users. It wasn’t exactly a success, the self-learning without a human filter leading the bot Tay to express racist sentiments.

Towards ever more content

AI is not synonymous with magic, however, but rather with data (and therefore content), mathematics, iterations and models.
While algorithms are certainly important for an effective conversation engine, it is the content which is truly essential. This content may take the form of text, images, videos or audio.

With the bot drawing its responses from a knowledge base, the quantity and quality of the data is essential in order to provide accurate and appropriate automated responses. This content may come from several sources. An incomplete knowledge base may provide a disappointing experience for the user due to incomplete or approximate answers, or even no answer at all.

Overcoming ambiguity

Conversational robots must also be able to “disambiguate” the question asked if the structure used by the user is unknown to the robot, or too vague. In this case, it makes suggestions to the user in order to more accurately determine the question or the scenario concerned.

This disambiguation phase depends both on the relevance of the underlying engine and on the work performed on the knowledge base. Certain requests may actually have several meanings. For example, for a request such as “transfer my contract”, does the user wish to change the name of the subscriber, or are they moving house?

It is essential to avoid misinterpretations or providing answers which are not sufficiently precise. Using mechanisms based on the concept of “emergent grammar” defined by Hopper, or on self-learning through collective intelligence (user clicks on the same suggestions following a misunderstanding), or on deep learning algorithms, engines can make relevant suggestions to content administrators to allow them to efficiently manage the knowledge bases of their chatbot.

Increasingly independent robots

We mainly use vertical chatbots, which offer a detailed level of knowledge on a specific but limited subject. By way of example, a chatbot dedicated to energy consumption will draw its replies from a database with tens of thousands of different formulations of questions. If Moore’s law is to be believed, the speed of information processing is doubled every 2 years. We can easily imagine that in the future, bots will be increasingly universal and independent, to the point of adapting to the subject and tone used according to the context.

Currently, knowledge must come from structured content, or else the automated structuring must be assisted by content administrators. Providing a very precise and appropriate answer to a question is often more complicated than it appears, when the answer needs to take into account parameters such as the user’s profile, their subscription or product type, the geographical location from which the question is asked, from which device… But inevitably, the machine will soon be able to construct its knowledge base alone, based on unstructured external content. It will therefore be capable of reading, understanding and structuring this content itself.

The question then arises of how AI is supervised. Will we legislate to establish the rules for the autonomy of AI and list prohibited fields of applications? Will it require constant supervision by humans?
The answer to this question is already known. Just as there is a law for telecommunications, energy or IT, so a “robot” or “artificial intelligence” legal category will come about, as Alain Bensoussan and Jérémy Bensoussan wrote about in their book “Droit des Robots” (Robot Law).

Do you Dream Up shares its experience, key success factors,  the different stages for a bot deployment project and how its technology works.