The big worry over big data

Last month, the Financial Conduct Authority (FCA) announced plans to regulate the use of big data within the insurance industry. Unsurprisingly, the news caused a stir in the sector, which has relied heavily on the use of big data in its operations for some time. Insights extracted from sources such as social media – which are already publically available – linked to other external/internal data sources are highly valuable to the retail industry when targeting customers through data insights.

In line with this proposal, the FCA recently published a Call for Inputs on the use of big data in the retail general insurance industry. The investigation will analyse the effect that the use of big data has on both insurance companies and the customer, with special attention paid to areas such as competition, consumer outcomes and regulation possibilities to meet both parties’ needs. During the Call for Inputs, the FCA will also be meeting with insurance companies, industry bodies and other interested parties, and then using the results of these discussions to determine whether a wider market study will be needed.

Big data, and its associated techniques of machine learning etc., often causes conflict within the financial services industry. Whilst other retail sectors utilise this type of information on a daily basis, the use of big data in insurance, in particular, triggers ethical worries amongst regulators – in part because this data comes from sources that include credit and debit card purchases, mobile apps and social media.

As a result, some argue that the use of this intelligence causes a breach in privacy rights, but there are a variety of adjustable settings in place to avoid this. For example, social media sites such as Twitter and Facebook provide their users with options to restrict private information from being published. Similarly, cookies require an individual’s permission to collect information regarding his or her browsing sessions, which means that insurers and other retail companies are only using information that has been made available by the customer.

One particular insight valued by insurers is supermarket habits, since ‘healthy’ purchases can sometimes be used to justify lower insurance premiums, based on the assumption that these shoppers will have a lower chance of illness. Of course, no one is in a better position to collect this type of information than the supermarkets themselves. As such, competition in this sector will be dramatically altered if more of the major supermarkets decided to enter the insurance sector, following in the footsteps of those such as the Cooperative.

The FCA has vowed to explore competitive advantages like these in order to determine whether or not they create an unfair environment for insurers. Organisations such as Facebook and Twitter, were they to enter the market, could also provide sizeable competition, since any company that is capable of analysing its own customer data for contextual insights would pose a serious threat to the current market leaders.

This type of information can also help with risk assessments, as companies could use this data to determine whether or not a claim is fraudulent. For example, if an individual has claimed for a broken leg in a car accident, insurers could check for any photos of a cast or mentions of physical activity on their social media pages or other data sources that will enable them to enrich their picture of the claimant.

When used in this way, big data can combat fraudulent claims, so that genuine claimants can be served more quickly and efficiently. However, savvy organisations could also use this information to improve their decision making processes, create bespoke products that cater for specific needs, and ultimately improve the customer experience.

As part of its review, the FCA will also be examining cases in which this information could be used to discriminate against those with atypical characteristics, such as those with unspent criminal convictions. As such, data privacy legislation will continue to be the Achilles heel of any attempts to mine publicly available data for behavioural insights, since some information will be deliberately withheld.

Insurers that want to avoid the wrath of the FCA will therefore need to conduct on-going due diligence when it comes to their composite data. Having easy access to Management Information (MI) that proves they are compliant will help firms address any ongoing complaints and also protect them from compensation claims, some of which may run into the billions.

Unless they have the right data and the correct culture to be committed to use this data for improving the customer experience as well as for their own protection – and an experienced partner that has the expertise to draw the right insights from it – insurers will leave themselves vulnerable in this area. As the FCA moves closer to a conclusion on big data in mid-2016, insurers will therefore need to find new ways of collating the unstructured and structured data available to them. If not, many will struggle to prove that they have used this data to act in the best interests of all of their customers.

Robin James, Product Evangelist, FusionExperience

Image Credit: Shutterstock/McIek