A typical use case for chatbots is answering repetitive questions. In the following, I show how a Swiss health insurance company is using the advantages of chatbots and now offers its customers a digital assistant.
The Swiss health insurance company SWICA has a wide range of products and a rather heterogeneous target group.
At SWICA, customer service and customer contact are our top priority. To increase customer satisfaction at the digital touchpoints, the company has decided to add an automated chat, i.e. a chatbot, as an additional communication channel.
For its first use case, SWICA’s online team deliberately focused on digitally savvy target groups who are already familiar with the medium of chat or chatbots.
Before users can chat with the chatbot, they receive a brief explanation of what a chatbot actually is. You will then be asked to accept the listed data protection provisions.
Figure: Swica chatbot “Data protection provisions at the beginning”
The chatbot then greets its users and introduces itself as the health insurance company’s new digital assistant. He says from the outset that he is still “new” and may not understand or be able to answer all user queries.
This is important so that user expectations are not set too high. If users have too high expectations of the bot’s capabilities, they are usually disappointed.
After the chatbot has introduced itself, it asks users in a polite and serious manner what topic they are interested in. Users also have the option of clicking on the “What can you do” button. The chatbot then shows which topics it can cover. If the user is interested in an insurance policy, the chatbot asks the user for further preferences and then shows him the insurance policies that are suitable for him. If the user wants more information on individual products, the chatbot goes into more depth and provides additional information on the individual insurance policies. The fact that the chatbot combines the use of different media is very advantageous here. In addition to text and buttons, it also uses images and other navigation elements. The images are particularly helpful when it comes to conveying the right emotions to users. Elements such as a checkbox questionnaire or carousel help the user to make a selection.
Figure: Swica chatbot “Use of different media”
The description of the individual insurance policies sometimes requires a lot of text. However, so that readers are not overwhelmed by too much text at once, most of the text blocks are divided into many small individual blocks. This makes it easier for users to follow the chat dialog.
Compared to some other chatbots, this chatbot is not a simple website guide, but actually takes the user right up to the point of taking out insurance or to the live chat or premium calculator. Other chatbots of this type often only refer users to the appropriate landing page, where they then have to read all the information again in classic text form. However, the SWICA bot displays the answers directly as text in the chat window. This avoids a media interruption and keeps users in the chat for as long as possible. Only when the user requests a quote does the chatbot open a new web window where the user can enter their information to create a quote.
Chat users also have the option of requesting a personal consultation. If this happens during customer service hours, a customer service employee will join the chat directly. The message “I would be happy to connect you with one of our specialists in this chat. Please be patient for a moment.”, the user is informed that the conversation with the chatbot will now end and a human advisor will join the chat instead. This means that the user does not have to leave the chat window. As soon as the next employee is free, they enter the live chat, introduce themselves with their real name and, thanks to a small color distinction, the conversation with the person is immediately recognizable. If such requests occur outside of business hours, the user can arrange a callback.
Insurance is usually seen by the target group as a very dry subject that is not much fun to deal with for long. This makes it all the more important that the chatbot does not annoy users, but guides them quickly and easily to their destination.
The health insurance company deliberately opted for a rather abstract name for the chatbot. It is called IQ. This makes it clear to all users right from the start that the person they are talking to is a chatbot.
Apart from that, however, the chatbot has no special character traits. It writes without much emotion and no particular personality traits of the chatbot stand out. IQ always answers very precisely and without filler words. Just like a computer.
With the chatbot, SWICA’s Digital Service Team wanted to create a touchpoint that is available 24/7. They have succeeded in doing so. Thanks to the round-the-clock service, they were also able to positively influence customer satisfaction and reduce the number of customer inquiries via other channels. Customer service employees now have more time again for complex inquiries or cross-selling and up-selling discussions.
At the same time, SWICA’s image has changed to that of a modern health insurance company that is always close to its customers.
SWICA’s chatbot is a successful example of chatbots in customer service, which answers the most frequently asked questions and at the same time advises users on the company’s offers.
Furthermore, this example shows very nicely that chatbots do not only have to contain text, but can also contain different elements such as images, buttons or carousel displays.
If you would also like to implement such a chatbot or another bot, I would be happy to help you with the concept work. Just get in touch with me. Click here to go to my contact page.
