Social media has earned a place in the life of almost every single person on the planet. There are now 1.97 billion active Facebook users as of April 2017.¹ As of June 2017, the world population was 7.5 billion, meaning a quarter of the world’s population is on social media.² Because of this, one would have expected insurers to start using this omnipresent technology much faster than is currently the case. But using social media for insurance is hard, and expert guidance is needed when navigating the technical, legal, and ethical landscape of social media analytics. In this chapter, I will explain some of the main challenges, and propose ways for how we as an industry can solve them. There will also be examples of insurers getting it right, and wrong, and two alternative scenarios for the future of insurance. Ready? Let’s get into it.

We can start by agreeing on the fact that social media is ubiquitous in almost all other industries. We use Facebook to login to our Spotify accounts and get music recommendations. We order flowers and groceries directly from our WhatsApp or WeChat accounts. And even banks have woken up to the massive opportunities inherent in analysing social media data, and use it for marketing and for helping users to a smoother customer journey. The birth of FinTech in the late 2000s has accelerated the development of social into financial services, and we are now starting to see the beginning of the same trends in insurance.³

When we started Digital Fineprint, an InsurTech company based in London, we identified three key challenges to using social media in the insurance industry, arising from the idiosyncratic nature of insurance. Understanding each of the three challenges is highly important for any industry player attempting to enter the space, so let’s go through them one by one.

First, insurers need to convince their customers to share their social media profile (often by clicking “Connect with LinkedIn” or “Log in with Facebook”). If there is not a clear customer benefit, users are less likely to do so, and insurers will not have sufficient data to analyse.

Secondly, the platforms themselves need to be aware and approve of the way insurers use this sensitive data. In a rapidly changing legal environment, platforms such as WeChat, Twitter, LinkedIn, and Facebook are all updating their privacy policies, and only the fastest moving technology companies are able to keep up with and stay ahead of the changes. Insurers lack the technical capability and expertise to deal with this very dynamic data landscape.

Finally, even if users and platforms agree to share data with an insurer, there is a challenging technical difficulty inherent in distilling insights from massive amounts of unstructured data. Predicting insurance needs, giving insights to insurance agents, and calculating risk classifications from social media data is extremely difficult, which is why even social media analytics companies working with banks have not yet moved over to insurance (as of the time of writing). It seems that only by focusing solely on insurance and solely on social media data can this technical challenge be overcome.

When Admiral launched “FirstCarQuote”, targeted at first-time drivers or car owners in November 2016, they were planning on using data from a user’s Facebook profile in order to set the price of their car insurance. This would be based on things like what posts they had “liked” and the way they spoke online – whether they wrote in short, concise sentences and used lists. Their launch failed, but three problems were painfully illustrated. First, customers saw no benefit in sharing their sensitive Facebook data with Admiral, and the insurer did not do a good job of communicating the benefit to the user. Secondly, Facebook’s developer agreements were not followed, which led to the access being revoked on the same day the initiative launched. Thirdly, media reports from several sources indicated that Admiral was not technically ready to analyse the relevant data, and thus would not have been able to make sense of any collected data. As a result, FirstCarQuote ended up being a PR failure and Admiral was forced by Facebook to pull the product less than two hours before it was due to launch.⁴ More insurers are now instead getting help from external vendors instead of building in-house solutions for using sensitive social media data.⁵

It is perhaps then no surprise that insurers have been slow to adapt to the social media revolution, despite a few attempts. In 2014, AXA announced a strategic partnership with Facebook and it was seen as a brilliant, innovative move by industry observers and insiders alike.⁶ But after almost three years, what has actually happened? Can customers buy insurance on Facebook yet? Is social media being used to augment underwriting data? Surprisingly, all the massive resources invested by insurance companies have not led to meaningful innovation in the niche field of using social media in insurance. This, I believe, is due to the three key challenges, which have yet to be overcome by the incumbent insurers. We therefore now see significant innovation and development from startups in the niche field of social media analytics for the insurance industry. One example is Common Easy, which allows users to create a “safety net” with their social network; Friendsurance, which uses an online P2P insurance combining social networks with well-established insurance companies; and Inspool, which uses social networks to let drivers connect with family and friends to form their own insurance groups.⁷

So what does all this mean for the future of insurance? I’ll take the daring approach and suggest two alternative futures for the industry, which will contrast two ways in which insurers can respond to the growth of social media. Let’s call them the “Skyscanner scenario” and the “Netflix scenario”. In short, I think that incumbent insurers will either end up as travel companies did in the mid-2000s, or, if things go well, as the Netflix of the 2010s. Let me explain how: when the travel industry started seeing online sales on comparison sites, direct-to-consumer plays and more, the industry as a whole closed their eyes and let innovative players scoop up the entire online market. Price became the key differentiator, and suddenly comparison sites became the key venue for booking especially airline tickets. Here, the industry missed out on an amazing opportunity to use social media data to give travel recommendations, build a trusted relationship, and create massive value for its customers. It is only now, ten years later and ten years too late, that the travel industry has started using, for example, Instagram to spread its message. The insurance industry is today, unfortunately, travelling on the same trajectory, with more and more comparison sites leading to commodification of personal lines of insurance.

Contrastingly, when Netflix started seeing its DVD rental business dwindle in the face of online streaming, they realized that by taking its most valuable resource – its database – it could completely turn around its fate. It therefore took its customer database and combined it with aggressive social media marketing, personal recommendations, and a great use of analytics.⁸ As a result, it became the leader in online streaming, and is now vastly more profitable than it could ever have been if it had relied on a DVD rental base. Insurers now have the exact same opportunity, and all they need to do is to take advantage of the innovative work being done in social media analytics. By combining customer insights with social media targeting of advertising, or by linking risk scores to online profiles, a whole new world of opportunity is there for the taking.

The use of social media in the insurance industry is inevitable. The data is too valuable and too useful for it not to be used by insurers. As an industry, we will simply have to agree on how, not if, it can be used in a way that benefits the customer. By overcoming the challenges already mentioned, we can take a first step towards a social media-enabled insurance company.

To have a look into the future, we can look to the east. Many industry observers have watched Chinese InsurTech ZhongAn grow from nothing since its inception in 2013, to offering 200 different insurance products and underwriting 3.6 billion policies for 369 million customers by the end of 2015.⁹ This is a growth journey that has absolutely astounded the industry. Observers are wondering what could have enabled it to grow so quickly, and already reach profitability.¹⁰ The author’s view is that everything so far discussed is represented in ZhongAn, as they managed to get everything right and overcome the three challenges. They took a “Netflix” approach to innovation, and created more lessons than any other InsurTech for the rest of the world to learn from. Incumbent Chinese insurers were left gaping in horror and in awe at the massive growth, and are now struggling to replicate the steps ZhongAn took to achieve its success.¹¹

So what did ZhongAn do differently to make all this happen? They integrated their entire insurance offering with WeChat, the largest Chinese social media platform.¹²