Chatbots are often a source of frustration. Here’s how we can improve them


Raise your hand if you’ve ever cursed, teased, or yelled at a chatbot. No surprise if you have any. These automated “assistants” – supposedly designed to make customer service smarter, faster, and more efficient – can certainly be a source of frustration for sentient beings.

Interactions with chatbots became more and more frequent in our daily life. But when we ask for information or try to resolve a problem, we are often annoyed when the chatbot cannot understand or misinterprets our request.

Worse yet, he advises us to contact the call center or visit a webpage, which defeats the original purpose of using chatbots.

There are two main reasons for negative user experiences. First, organizations often present the chatbot as too “human”, leading to Unrealistic expectations on the chatbot’s ability to understand human language, including questions and nuanced commands.

Second, many chatbots are rule-based and have a narrow knowledge base, which means grammatical and syntactic errors can confuse them, and complex questions often cannot be answered, disappointing customers.

A two-way street

While it’s easy to blame the chatbot for a miserable experience, we have to realize that just as it takes two hands to clap, it takes both the chatbot and the customer to create a satisfying interaction.

While previous studies have mainly focused on the chatbot, including why companies implement them and the design cues that characterize them, the role of the customer in these interactions has not been given much consideration.

In our latest research, we focus on how customers treat chatbots and provide ways to improve the experience.

We find that in order to create constructive and meaningful engagement with a chatbot, the customer’s actions and feedback and the willingness to make it work are as important as the chatbot’s own functionality.

Understanding chatbots

We have identified six distinct types of human-chatbot interactions: socialize, collaborate, challenge, accommodate, engage, and redirect.

These vary depending on who is leading the conversation (the chatbot or the customer), the “reality” they perceive from each other, their social cues, and the customer’s effort.

In the case of socialization, the chatbot tries to entertain the customer, for example by telling them jokes or trying to cheer them up if they detect a bad mood.

Collaborative interactions are those conversations where the chatbot and the customer work together on customer needs, such as booking a flight or understanding the root cause of a problem and identifying solutions.

Socialization and collaboration interactions involve fluid exchanges between the chatbot and the customer and generally lead to positive results.

‘What is the meaning of life?’

Accommodating interactions are those where the customer is in the driver’s seat, helping the chatbot understand their needs by changing the way they phrase the question or statement, repeating their request, or clarifying their intention.

On the other hand, an engaging interaction sees the chatbot more engaged than the customer, trying to provide an answer to a question or resolve a customer’s problem.

In these cases, chatbots often ask follow-up questions and provide additional information that might be relevant. These two types of interactions, however, often leave customers without the required information.

In some cases, people see the novelty of chatbots as an open invitation to challenge them and see when it breaks. This type of interaction usually doesn’t get anywhere, as most chatbots aren’t trained for off-topic questions like “will you marry me?” or “what is the meaning of life?” “.

Finally, when redirecting a customer, chatbots act more like a browser, pointing to alternative sources of information such as the company’s website, and do not respond directly to inquiries. These interactions are very short and may not be an ideal outcome for the client.

Three keys to success

Based on our research, we offer three tips for your next chatbot encounter:

  • Remember that a chatbot is not human and many chatbots cannot understand nuanced natural language, so try not to use complex sentences or provide too much information at once;
  • Don’t give up too quickly – if the chatbot doesn’t understand your question or asks the first time around, try using keywords, menu buttons (if available) or short sentences; and
  • Give it a second chance: Chatbots learn new “skills” over time.

Organizational advice

The introduction of chatbots has redefined the way customers, employees, and technology interact, and we’re encouraging organizations to take a holistic view of their customer service systems when redesigning them.

Special attention should be paid to the changing role of customer service employees who have to work with chatbots. In addition, we recommend that organizations:

  • Reinvent a customer service team: involve people in the overhaul of customer service delivery through a mix of chatbots and real employees;
  • Treat chatbots like a new (digital) employee: spend time and effort expanding their skills;
  • Find the right place to send a request to a contact center employee. Some chatbots refer people too early (causing congestion), while others frustratingly offer the option late. Experiment to find the right time; and
  • Monitor chat interactions. Find out how and what questions customers are asking and expand your chatbot’s knowledge base accordingly.

The authors acknowledge Thai Ha Nguyen’s contribution in the preparation of this article and the original review article on which it is based.The conversation

This article is republished from The conversation under a Creative Commons license. Read it original article.


Comments are closed.