Dear Visitor, this website uses permanent cookies to ensure data privacy as well as to provide access to special content and technical functionalities. With your consent, we may also use cookies for website analysis. For details and how to give or revoke your consent, read our Data Privacy Statement.

  • EN
Top navigation or filters may restrict results of your search. To see all possible results click here.

The uptake of chatbots in insurance

31 May 2022
Customers are driving insurers to engage more with AI-powered automated messaging technology like chatbots. Johannes Seebacher, co-founder of software provider riskine, describes their benefits and how they can even be used to offer risk analysis for customers.

AI-powered chatbots will soon be the norm for insurers as more and more companies deploy them to handle more than just simple questions from customers. They are also used to provide wide-ranging financial advice and for claims handling: 7% of insurance customers say they trust chatbots when making a claim.

Another comprehensive survey shows that while consumers can be wary of chatbots, customer satisfaction actually increases for those using them: Analysis by Freshdesk of 107 million customer service interactions at 4,500 companies reveals that customer satisfaction scores increase by 7% when bots are used.

Quality of request and advice is key

That said, first impressions are vital. “Chatbot providers must ensure that service and information quality is good, as poor initial experiences can create doubts, resulting in loss of trust,” reports Himanshu Joshi of the International Management Institute in New Delhi, India, the world’s second-largest country by population. “Professional interactions, quality of request and advice, ensuring privacy et cetera can help in building trust,” he writes in his 2021 paper “Perception and adoption of customer service chatbots among millennials”.

Giving advice is what the insurtech riskine does, and its chatbots can do much more than just answer simple questions: “We provide software to digitally advise insurers’ customers – our sole focus is good advice,” says Johannes Seebacher, co-founder and CTO of riskine. The company was founded in 2016 as a B2B fintech creating digital advisory solutions based on artificial intelligence, including chatbots.

Examples of current use cases

Answering customers’ frequently asked questions is a compelling use case for a chatbot. But Seebacher sees chatbots going further, giving advice to financial customers when buying a product, including talking about the customer’s risks as well as making financial projections for investments.

One German digital insurer using riskine’s open source insurance data model identified additional areas where the company’s chatbot technology could add value. “One of the areas was the Notice of Loss process, where their goal was to increase automation and customer satisfaction,” says Seebacher.

riskine is also working with an Austrian household insurance company to answer questions from users and guide them to the relevant resources. “In the case where the customer wants to buy an insurance policy, the bot would take the user through an advisory conversation including real-time quotation,” says Seebacher. “The goal is to take the user as close to the decision as possible, while giving them quality advice. Once the user is ready, they are routed to the end of the usual quotation journey, where all the relevant data is already prefilled.”

Greater accuracy through domain-specific models

In order to avoid some of the problems associated with generalist chatbots, providers like riskine are banking on sector-specific chatbots built on proprietary natural language processing systems. 

Natural language processing

Natural language processing systems allow users to ask questions in their own language and in whatever form they choose. An AI-powered engine translates that into something the computer can recognise as a question for which it has an answer.

“If you go to Google, which offers open source natural language processing through an Application Processing Interface or API, you get this general engine and you have to train it yourself,” says Seebacher. “It is just not possible to go into all the depth needed. You will always have this general model that will cut corners. With deeper domain-specific models, you can do things that are otherwise not possible.”

“Since we offer the natural language intelligence out of the box, each freshly set up bot already has a full understanding of insurance and financial contexts and can also put that together in a coherent conversation.”

The future development of chatbots

With both insurers and younger customers keen to embrace chatbots, the technology is here to stay – but it may morph into something different: Seebacher is convinced that the chatbots we see today, typically tucked away in the corner of a website, will be replaced by something more powerful. He says, “There is huge potential for bots embedded into customer chat solutions where users can chat to a human agent where the agent is prompted by a bot to automate some of the conversation.”

“We think that the main benefit from this is not the time savings but the increased quality of advice, since the human agent is able to profit from the bot’s knowledge,” Seebacher says. “Human interaction is still central in the financial/insurance advice industry,” he adds. “There might be important financial decisions that the end customer wants to discuss with a human professional. On the other hand, it is important for our customers to stay in touch with their customers.” And chatbots can provide them with additional relevant touchpoints.

Related solutions

User questions

Answered questions


Unanswered questions


Thank you for your question! We will inform you when the page owner answers your question.

Please note that submitted questions are public.

Views: 1103

Downloads: 0

Ratings by number of stars:
0 %
0 %
0 %
0 %
0 %

Page is favored by 0 user.

Contact inquiries: 0