Today Fabian Reinkemeier talks to Sophie about the use of conversational user interfaces. Find out what it is, why it is so important and how you can use it successfully in this podcast episode.

About Fabian Reinkemeier

Fabian Reinkemeier works at elaboratum, a digital consultancy specializing in customer journeys, strategy, design and implementation. His focus is on e-commerce topics and he heads the Conversational User Interfaces task force. He is also doing his doctorate at the University of Göttingen as an external doctoral candidate, where he is researching the interface between marketing, IT and psychology.

We need to talk about this

As an attentive listener of Sophie’s podcasts, Fabian contacted Sophie independently, as the topic of conversational user interfaces is particularly close to his heart and has not yet been covered much. Conversational user interfaces are user interfaces that are characterized by the fact that the user is supported by a natural language dialog. This can be done either rule-based or through automatic intents, i.e. with the help of AI. Fabian’s goal is dialog-based interfaces that use natural language. In recent years, people have become far too conditioned to interact with graphical user interfaces, he says.

Voice commerce is all the rage

A significant increase in voice commerce is expected over the next few years. This is because the technology uses natural language on the one hand and combines the advantages of online trading, such as 24/7 accessibility, on the other. In this way, customer concerns can be resolved quickly and easily and the user is given the feeling of a real consultation.

Typical use cases and the dilemma of disappointed customer expectations

A survey of 1,000 respondents revealed that the most common use case for a bot is customer service, at 42 percent. This result is hardly surprising, as customer service processes can often be automated quickly and easily, thus saving money. What is much more interesting, however, is that 72 percent of respondents also stated that their expectations were not met by the bot. This is partly due to poorly implemented chatbots, says Fabian Reinkemeier, but there is another important factor that many people often forget. There are often situations (e.g. frustrating situations) where customers simply don’t want to talk to a machine, especially if it doesn’t understand them. In the survey, many customers stated that they would have preferred a chatbot when searching for or ordering products. Although there are many different use cases, the primary goal should always be to meet customer expectations.

Discrepancy between theory and reality

It can often be observed that there is a large discrepancy between the information provided by users in surveys and their actual behavior in reality, or what they do with a chatbot on a website. According to Fabian, this is because many people are not aware of how they actually make decisions. In addition to an intuitive, instinctive decision-making system, humans also have a rational system. Even if many people don’t like to hear it, studies have shown that 95 percent of human decisions are made subconsciously. However, chatbot optimizations are often aimed at addressing the rational system and therefore miss the mark.

The path to the right use case

The key to success is to focus on the user and their behavior. Those who understand customer decisions and know how they are formed are the winners, says Fabian Reinkemeier. Such an approach also offers the potential to influence customer decisions. Focus groups, for example, can be used to better understand customer decisions. These make it possible to specifically understand how customers would search for a product via a chatbot or seek advice, for example. The bot can then be designed or optimized based on the results with the customer at the center.

This approach also prevents customers from leaving the website prematurely. The website is available 24/7 and you can store from the comfort of your sofa. Nevertheless, the extensive product range of some online stores overwhelms many users. Because, says Fabian, a complex product range means that a customer has to make a number of decisions for or against a product. Some users withdraw from this and leave the website prematurely. In contrast, a sales assistant in the store is there to help and advise the customer and tries to find out what the customer’s needs are by asking clever questions combined with empathetic behavior. Combining the advantages of a location-independent shopping experience with the natural language used by a salesperson is where the benefits of conversational user interfaces lie. At this point it must of course be mentioned that online shopping is only an example use case, many others are possible.

Learnings

1. the use case must be clearly defined. What is the advantage of using a chatbot? Wouldn’t a better search or better filtering perhaps also help?

2. puts the user at the center of the bot’s design. Try to understand the customer’s decision. Focus groups can provide support here.

3. a chatbot is never finished. Get feedback from customers and test the chatbot extensively. In this way you increase the quality of the bot and increase the traffic on your website.

4 When optimizing the bot, make sure that the optimizations are aimed at the intuitive, instinctive decision-making system. We make 95 percent of our decisions in this way.

It’s best to listen to the podcast episode with Sophie Hundertmark and Fabian Reinkemeier pure. Have fun!

And if you don’t want to miss any more podcast episodes, then sign up for the mailing list at the bottom of this page.