Experts agree that the future of insurance is digital. Today s customers are used to receiving services, goods and information anytime and anywhere at the touch of a fingertip - and expect the same from insurance companies. But the industry is still lagging behind.
You are driving on the motorway when, in the middle of roadworks of all places, the man behind you gets too close and hits your bumper. Luckily, its just a minor accident. The party causing the damage quickly takes a few photos by smart phone and sends them together with your contact details to his motor insurer. A few minutes later, the system has already roughly estimated the costs and authorizes the repair in a workshop of your choice - without police involvement, long traffic jams on the motorway and without the need for specialized expertise. You receive an e-mail with the reference number under which the damage will be processed. A photo of the invoice a few days later will suffice and you receive payment for any damages within 24 hours.
Futuristic dreams? Not at all. In pilot projects such scenarios are already possible through the use of smart software. And the potential is enormous - especially in a data- and information-driven industry like insurance.
But the reality is different. Paper-based processes are the rule rather than the exception and much is still done manually. This is time-consuming and ties up employees who could otherwise perform other tasks.
There are many reasons for the low level of digitization in the insurance industry:
First, life insurance, health insurance and property and casualty insurance are often operated by legally independent companies. As a result, a customer within a group is managed in different IT systems, some of which are unable to exchange data with each other.
Secondly, cost and competitive pressure in recent years has led to smaller insurance companies being bought up by the big ones. At the end of 2017, 436 companies were present on the United Kingdom domestic insurance market. That is a decrease of 237 companies compared with 2004. Here too, the result is a patchwork of different IT software and hardware with many interfaces that are incompatible. Obsolete IT systems from the 1990s make system integration even more difficult.
Thirdly, established insurers are struggling with rigid ways of thinking and internal competition. Large companies like AIG or AXA find it much more difficult to simultaneously offer digital direct insurance in direct competition with their own sales channels. Long decision making processes and rigid hierarchies do the rest to stall the introduction of digital solutions.
As a result, very few insurers have modern IT systems that allow customer data to be bundled over the entire contract term and all interfaces. This results in a deficit of high-quality and correctly classified data to train algorithms.
As an intermediate step, many insurance companies are opting instead for robotic process automation (RPA). In contrast to classical automation in production plants, here virtual robots are used instead of physical robots, which previously perform manual, repetitive tasks independently. The software robots imitate human activities on the computer; complex system integration is not necessary.
The advantages of this technology are obvious: the virtual workforce can be easily integrated into existing IT landscapes without prior programming knowledge, the processes are faster and more error-free and costs are reduced. According to a study by consulting firm Capgemini, over 40% of companies already use RPA.
RPA is particularly suitable for frequently recurring, structured activities that obey fixed rules. For example, software robots can accelerate the migration of large amounts of data, the conclusion of new contracts or the adjustment of existing policies.
However, software robots work exclusively deterministically and are limited to simple work processes. They are therefore not in a position to react flexibly to deviations, for example if data is entered incompletely or incorrectly.
The use of artificial intelligence is much more promising here. AI refers to computers that are able to solve problems more or less independently and with more or less assistance and learn from experience without the necessity for predefined rules.
Artificial intelligence, for example, can do much more than speed up insurance processes such as the conclusion of policies or the settlement of claims. It will lead both to a more accurate assessment of risks and to more personalized products. The better insurers know about their customers, the better they can refine risk profiles, adjust prices and tailor insurance packages.
AI, in combination with sensors and intelligent devices, can help to prevent such risks. In industry, machines can be repaired before expensive downtime occurs in production and manufacturing. Telematics tariffs in motor insurance are already moving in this direction by promoting prudent driving style. Technical means such as a box equipped with sensors or a smart phone app record the driving behavior of the car owner. In addition to braking behaviour, speed and acceleration, other factors such as location and time are also evaluated. If you drive carefully, you get discounts on car insurance.
There is still a long way to go before established insurers will be able to take this broad-based approach. Insurtechs have a clear lead here. On the one hand, they see themselves as challengers and consistently think of insurance companies from the customers point of view. On the other hand, they have no legacy IT structures or hierarchies, but are building their company on a greenfield site.
Both of these factors help Insurtechs to better meet the needs of today s insurance customers for simple, flexible and, above all, digitization, than established companies. Robot-assisted process automation is therefore only the first step, but it must not stop there. IT systems urgently need to be modernized and the divisional approach must be overcome. In the long term, it is only those insurance companies in the market that manage to record and evaluate customer data in a structured way over the entire customer journey that will be successful.