Corporate Advisory

Behavioural patterns to influence price modeling

One of the biggest insurance companies in Britain is to use social media to analyse the personalities of car owners and set the price of their insurance.

The unprecedented move highlights the start of a new era for how companies use online personal data and will start a debate about privacy.

Admiral Insurance will analyse the Facebook accounts of first-time car owners to look for personality traits that are linked to safe driving.

In contrast, evidence that the Facebook user might be overconfident – such as the use of exclamation marks and the frequent use of “always” or “never” rather than “maybe” – will count against them.

The initiative is called firstcarquote and was officially meant to launch this week but that was delayed at the last minute on Tuesday night. It is aimed at first-time drivers or owners – although anyone with a licence can apply. The scheme is voluntary, and will only offer discounts rather than price increases, which could be worth up to £350 a year. However, Admiral has not ruled out expanding firstcarquote.

The rapid growth of social media and personal technology has given insurance companies and employers swaths of data they can access to analyse customers or employees. As well as Admiral’s car insurance scheme, insurers are looking at how they can use the rise of smartwatches and fitness trackers to monitor people’s health.

The scheme is based around algorithms that have been developed by Admiral. The technology uses social data to make a personality assessment and then, judging against real claims data, analyse the risk of insuring the driver.

Yossi Borenstein, the principal data scientist on firstcarquote, said its algorithm looked for correlations between social media data and actual claims data. The technology will evolve as firstcarquote attracts customers and gathers more evidence about the correlations, meaning the importance of items identified on social media could change.

Borenstein insisted that Admiral would not have access to information about what its customers look at on Facebook or what their friends do. The company would only have access to the information gathered during the quote process and would have no ongoing access.

“If this is successful, it could be revolutionary,” he said. “It could be truly transformational.”

See full article here