Get All Access for $5/mo

Why Chatbots Failed and the Future of Conversational AI How can companies use this technology to create relationships with customers and increase sales?

By Maja Schaefer

Opinions expressed by Entrepreneur contributors are their own.

You're reading Entrepreneur Europe, an international franchise of Entrepreneur Media.

ipopba | Getty Images

The pandemic has made every business digital by default, but even prior to Covid-19, businesses recognised the potential of conversational artificial intelligence (AI).

One of the first uses of AI in customer service was in chatbots. These days, "chatbot" almost sounds like a dirty word. It comes up again and again in my conversations with other founders: chatbots have developed a reputation for providing inefficient, impersonal, and frustrating customer service, which are the exact problems that they were meant to solve.

What went wrong? There are three main reasons why the first chatbots failed:

They didn't use machine-learning effectively. AI is meant to learn over time. The bot should be able to learn from past interactions and improve itself. Early chatbots were simply not as advanced as they are today.

They weren't kept up to date. Even now, some businesses try to build their own chatbots but fail to manage them properly. For example, they don't keep their chatbot updated with new product information, which leads to frustrated customers and abandoned sales.

We tried to automate everything. It's important to learn what to automate—and what not to automate. More complex questions should be left to human agents. Sometimes, you just want to talk to a human and if companies want to provide their customers with an amazing customer experience, we need to understand this and set up our processes accordingly.

Persistent misconceptions.

Chatbots are a means to an end, not the end goal itself.

Misunderstandings about this are still common. I hear it from founders and customer support leaders and teams alike: they're worried that chatbots will replace their customer service teams and dehumanise the customer experience. On the other hand, many are asking their chatbot to do too much.

Businesses need to learn from these previous mistakes around how chatbots work, what they do, and when to use them.

New solutions.

So, how can companies use this technology to create relationships with customers and increase sales?

Use the best technology available. Machine-learning technologies are much more complex and effective now. I always advise companies to check how easy it is to maintain the conversational AI software that they're considering. Implementation is the exciting part. In my experience, the most frequent challenges come afterward, when it's about maintenance.

Know what (not) to automate. We've seen that emotional situations—for example, with a customer who is stressed or angry—are much better handled by a human agent. Other requests don't need this same high level of empathy. Customer requests like changing a shipping address or upgrading a subscription can be easily handled by conversational AI. As a machine-learning community, we're constantly improving and will increasingly be able to automate more situations.

Forget the stereotypes. In my experience, CSAT (Customer Satisfaction) increases after the implementation of automation. Customers quickly get the answers they want and those who have more complicated issues get the help they need. Customer service agents are less overwhelmed by repetitive work, so they have the bandwidth to be even more attentive and patient. Conversational AI improves the customer experience—which is what customers actually want.

What lies ahead.

There's a huge market opportunity: globally, 85 percent of consumers would like to be able to message with brands. But companies need to get it right. Just one poor support experience is enough to drive 50 percent of customers to a competitor.

Europe is leading the way in how we perceive and use AI. In 2020, the team behind Denmark's emergency helplines was supported by chatbots, which helped with Covid-19 management. Similarly, ALAB Laboratoria in Poland became the official government partner for Covid-19 testing and used chatbot integrations to scale their appointment booking operations. And in April 2021, the European Commission proposed new actions and rules for trustworthy AI.

With the help of conversational AI, human customer support teams can spend more time on meaningful work, like nurturing client relationships and generating revenue. Plus, this helps with staff retention. The most common reasons for staff turnover in customer support teams that I've seen are burnout and a lack of opportunities for growth.

Don't be put off by old ideas about conversational AI—the right tools will be able to provide the right support. And this is the right moment: chatbots may have failed in the past, but they won't fail you now.

Maja Schaefer

CEO & co-founder, Zowie

Leadership

Visionaries or Vague Promises? Why Companies Fail Without Leaders Who See Beyond the Bottom Line

Visionary leaders turn bold ideas into lasting impact by building resilience, clarity and future-ready teams.

Science & Technology

5 Automation Strategies Every Small Business Should Follow

It's time we make IT automation work for us: streamline processes, boost efficiency and drive growth with the right tools and strategy.

Business News

Former Steve Jobs Intern Says This Is How He Would Have Approached AI

The former intern is now the CEO of AI and data company DataStax.

Business News

SpaceX Sent a Toy Banana Into Space. Now You Can Buy One — But Not in Time for the Holidays.

Starship's sixth test flight carried a payload, a fake banana, as a zero-gravity indicator.

Marketing

5 Critical Mistakes to Avoid When Giving a Presentation

Are you tired of enduring dull presentations? Over the years, I have compiled a list of common presentation mistakes and how to avoid them. Here are my top five tips.

Science & Technology

Why Businesses Are Relying on Automation to Survive the Labor Crisis

Robots are revolutionizing industries by addressing labor shortages and enhancing efficiency, while businesses navigate challenges like workforce adaptation and high implementation costs.