Adopting AI for insurtech: how does the German ecosystem compare?

by | Feb 7, 2020 | Blogposts | 0 comments

In early 2019 I joined Tractable, a UK tech company using AI to analyse auto damage, to lead its expansion efforts in Germany.
Tractable is one of the largest and most developed insurtech companies in the UK. We’ve raised $30m from leading US VCs to scale our solution globally, and insurers across the world are using our AI to accelerate their auto claims handling. But how does the German insurance ecosystem compare when it comes to adopting AI? Here’s what I’ve learned so far:

1.German insurers are hesitant to trust AI

Tractable uses computer vision – the process of a machine understanding what’s in an image – to analyse auto damage. This speeds up how quickly claims can be resolved, for example by helping assessors take decisions (about, for example, total loss), resulting in substantial cost savings. In the UK, Tractable even helps leading insurers generate estimates of end-to-end damage with its AI, often without human correction. 
In Germany, though, there’s a problem. Most German insurers have already tried other AI solutions, and been hugely disappointed at what they’ve seen – which means there’s a lot of scepticism to overcome. Essentially, from experience, Germans don’t believe this tech works – and that means getting those initial conversations can be difficult. 
Fortunately, we have ways around this. For one, our AI is the global market leader – it’s been trained on many millions of images of auto damage – and it’s already in play with some of the world’s leading insurers, across 9 countries. That’s an asset when it comes to challenging preconceptions, but you also have to show that the tech works. We’ve been extremely active in providing live demos of how the AI can break down a photo of auto damage, analyse it and suggest next steps. Live testing helps prove the AI’s power and its possibilities, and demonstrates to skeptics that AI isn’t something that’s a few years away – it’s here, and we can show you that it works now. 

2. Large, helpful, insurtech ecosystem

Germany is, of course, a key target market for Tractable. German insurance companies are some of the most established in the world – and they live in an extremely competitive market. That means there’s a lot of opportunities on the table, especially as we believe we can save insurers as much as 1 point on their combined ratios. But to consider working with you, German insurers also need to know that your company is serious and for the long-term – not least as they need to differentiate between many start-ups from around the world, all touting new solutions that may or may not help (or even work). So, it was critical that we established a footprint in the country quickly, and that we met with as many of the major players as possible. Fortunately, Germany has a well-developed insurtech scene, including hubs in Cologne and Munich. We were able to leverage the connections of bodies and networks such as Plug and Play, DIA and Insurtech Lab to get the introductions and endorsements we needed. 
It also helps that we have already worked in other countries with subsidiaries of major German insurers. While the industry can of course be slow moving, that – combined with other examples of how we can save insurers money, and create increased satisfaction amongst policyholders – has spurred more conversations, and sped up the decision-making process. 

3. Finding the right fit is paramount, especially in Germany

Another challenge is that not all markets are the same, and the fit for our product in Germany is different from elsewhere. In particular, the German market has many independent appraisers and appraisal companies, which is unusual and can make it difficult for an insurtech to scale. (For example – in contrast, in Austria insurers actively approach us, because they see how they can use our tech to assess pre-damage when assigning new policies for used cars.) 
However, a positive that we’ve discovered within Germany is that there is a well-accepted process of exchanging data early on in order to carry out initial testing – almost like a ‘light’ proof of concept. That means we can find out exactly where our product might fit with German insurers quickly. In addition, what’s happened for us with other customers worldwide also helps: as to date, every insurer that has trialled our solution to a serious extent has found value in it and become a customer – so getting commitments to that early testing can be a win for us. 

So, with these learnings under our belt, what happens next?

Well, we’re engaged with exciting discussions now with German insurers, with the aim of finding exactly the right use case so we can start working with them in 2020. And once we’ve secured those deals, we’ll look to start scaling our presence further in the country, as we have elsewhere in the world. For example, we’ve already ramped up the number of staff we have on the ground in Tokyo from 0 at the start of 2019 to 10 today, in order to keep up with demand. 

But for me, perhaps the biggest win will be changing how German insurers think About computer vision and how it can help them. I believe our solution will be the catalyst for a new way of processing auto claims in Germany, as it is beginning to be in France, Poland, the UK, Japan and elsewhere too. I look forward to helping the ecosystem see that AI is not for the future – it’s here, and ready to make a difference, today.