Insurance-Canada.ca - Where Insurance & Technology Meet

Social Media, Analytics and Insurance: Managing Disparate Data for Profit and Growth

According to a recent white paper sponsored by Tata Consultancy Services, insurers can benefit significantly from social media by implementing strategies involving the use of predictive analytics, but have numerous hurdles to overcome before benefits can be realized.

The paper, “Predictive Analytics Enables Insurers to Drive Profitability and Growth”, notes that “social channels like Facebook, Twitter, and YouTube provide access to an enormous number of existing and prospective customers.”  The potential benefits include increased sales, better profitability per customer, reduced expenses, and better control over claims.  However, these benefits can only accrue with effective use of data driven by the social media channel.

Frank Diana, Director of Business Analytics, Global Consulting practice, Tata Consulting Services explains, “Data challenges become more complex as insurers look at adopting strategies around social media and strategies around incorporating social technologies inside their enterprise.”

Social media will drive large amounts of structured and unstructured data.  “Managing this data becomes a real, practical challenge. And unfortunately, while the quantity of data is high, the data quality is often low.”

The white paper suggests that the use of predictive analytics is necessary to transform the data into actionable information. “Predictive analytics differs from traditional business intelligence in that it uses mathematical models to gain insights from data and develop forward-looking conclusions. Predictive analytics enables insurers to better understand future events and behavior by examining the available history of current and prospective customers.”

This is a non-trivial exercise for many insurers for a variety of reasons.  Legacy systems are an obvious impediment.  As significant, according to the white paper, is “corporate culture and organizational structure in which people are not accustomed to doing things differently and are slow to change.”

For example, many companies are product focused, rather than customer focused.  By definition, predictive analytics and modelling cuts across product lines.  (We noted some additional challenges with introducing predictive modeling in systems replacement projects in a post in August.)

Overcoming the challenges is non-trivial, but offers substantial rewards in combining the separate disciplines of marketing, customer service, claims management, and underwriting, “linking underwriting and claims enables insurers to maximize insights on customer behavior,” the paper notes.