How insurers can turn machine learning from a threat to an opportunity
The new oil
It has become something of a cliché to describe data as “the new oil” of modern digital business. The truth is that, for insurers, it has been a vital driver of success for decades. Data is the key to unlocking insight into customer behavior and claims. From this data can be deduced patterns which enable insurers to better assess risk, develop policies and price their products more accurately.
The advent of machine learning offers a new opportunity to supercharge this approach. This sub-discipline of AI is focused around applications which have the ability to learn for themselves over time. In so doing, they can find hidden patterns in data to generate new insights for insurers, improving the accuracy of pricing and risk assessments. According to KPMG, machine learning also has the potential to “greatly simplify onboarding and customer service, in particular the claims settlement experience.”
There’s just one problem: the default setting of the insurance industry, quite understandably, is to stick with the familiar. This means working with historical proprietary data sets. But while applying machine learning algorithms to this type of data will provide marginal gains, it’s not the rocket fuel that could be driving stellar advances for the industry.
Time to join a revolution
According to Accenture, 75% of insurance executives agree that AI will revolutionize the industry. But while adoption of these tools is growing, there’s a real danger that insurers could be missing out. Instead of using historical data, they need to find alternative stores of real-time, third-party data to drive innovation-fuelled growth. This is the kind of data on which the likes of Amazon, Google and Facebook have built their fortunes. By applying machine learning to this info, innovators like these are able to serve up compelling products their customers didn’t even know they wanted, personalized search results and advertising, must-read content, and much more.
In fact, the data that many of the largest tech firms process on their customers today could make them prime candidates to enter the insurance space themselves. This should be a wake-up call to the industry: respond by turning this threat to your advantage and you could drive significant competitive gains.
Teaming for success
How can insurers do this? By partnering with third-party sites as well as other firms whose business model leverages real-time data and analytics. The smart home is a great example. Modern gadgets produce data constantly as they communicate with the world around them. By accessing this information, insurers could reduce premiums and expedite claims by monitoring temperature levels or smart security systems, for example.
In another scenario, an insurance firm partners with a shared workspace provider. By accessing the latter’s anonymized, real-time data, the insurer is able to offer small commercial risks more dynamic, personalized pricing. This approach also provides hassle-free customer on-boarding to the workspace provider, saving its business customers from a normally lengthy application process.
The same model could be applied to numerous sectors, from travel sites to logistics companies, payments providers, online property firms, mobility businesses, retailers and gig economy platforms. It’s a win-win for both insurer and its partner: driving growth on both sides by enabling online providers to add value for their customers by offering simpler, smarter insurance.
Getting there will require insurers to think very differently about how they view data: transitioning from a historic, closed approach to a dynamic, partner-led model. But it’s an opportunity waiting to be grasped by those with the ambition and appetite. If not, others will be ready and waiting to step in.
The Coronavirus, and its drastic changes on society, has changed how people
Customers want selection when shopping online; it’s where we go to find
We’re all spending more and more of our lives online. This means