It is estimated that the world generated 1,200 billion gigabytes last year alone. How does one draw insights from such a vast amount of data to make actionable decisions? Can insurers use this overwhelming amount of data to improve their overall profitability and efficiency?

Availability of large volumes of consumer behavioral data, advances in device and sensor technologies, improvements in predictive analytics and simulation techniques, and methods for visualizing large volumes of data are revolutionizing how P&C insurers consume large sources of internal and external data to and make business decisions.

Let’s look at the drivers of predictive analytics and how insurers can use certain techniques to maximize the benefits of this technology.

Why Now?

Focusing on future projections is at the heart of every insurer’s business. Whether they are P&C or life and health (L&H), they take cash from their customers today and make a promise to pay for certain eventualities or future conditions. The “future” varies from a couple of years for personal lines insurance to 30 years or more for life insurance and annuities. Insurers’ actuarial and underwriting divisions have always used predictive modeling in one form or another. However, predictive analytics have recently gained ground in claims, as well as marketing, sales and distribution.

A number of advances over the past decade have made it not only possible but essential for insurers to explore broader applications of predictive analytics:

  • Accelerating technology and consumer adoption. Advances in sensor, computing, and communication technologies are adding to the volume of data and enabling the analysis, interpretation, and visualization of it. The rapid growth of mobility and mobile data use, coupled with social networking, has resulted in exponential growth of behavioral data. The “Internet of things,” which makes devices from automobiles and plant machinery to dishwashers and washing machines relay real-time data, substantially increases the amount of information generated. Smart sensors that cost just a few dollars can measure a single attribute, such as temperature or moisture, and communicate the readings on a real-time basis. As these technologies improve and their use accelerates, they will increase the amount of available information.
  • Increased data availability. Consumer adoption of rapidly evolving technology and automation has resulted in the generation of billions of gigabytes of data by insurers, government organizations, regulators, non-profit organizations, rating agencies, and independent data aggregators. By some estimates, the production of data has increased from 150 billion gigabytes in 2005 to 1,200 billion gigabytes in 2010. As insurers have started integrating their legacy systems, externalizing data into enterprise data warehouses, and developing a single view of the customer, their internal company data sources have significantly improved in terms of quality and quantity. Moreover, external data sources have become more standardized, allowing for greater sharing of data. Non-profit data aggregators and for-profit data providers have facilitated the flow of information between different parties.
  • Sophisticated analytical techniques. The need to generate faster and better insights from increasing multi-media data is resulting in fresh techniques for analyzing text, speech, video, and sentiments. Analysis of online interactions, unstructured text, speech, video, and social data mining have all emerged as distinct areas of focus.

Value to Insurers

Professionals are using predictive analytics in underwriting, pricing, marketing, sales, distribution, customer service, claims, reserving, and hedging. While many personal lines carriers have seen lower profitability as a result of price-based competition, early adopters of predictive analytics in the personal lines sector have managed to price risk better and attract more profitable customers. As a result, companies that are not considering implementing some kind of predictive analytics will be at a disadvantage.

Generally speaking, there are four application areas for predictive analytics:

  1. Claims management. According to the Insurance Information Institute, fraud—most commonly, staged accidents and claims padding—costs P&C insurers more than $30 billion annually. By analyzing historical claims information and demographic profiles, predictive models can identify potential fraudulent cases for further investigation. This allows claims adjusters to focus on suspicious cases and conduct more detailed investigations. Predictive analytics can also reduce losses. By analyzing the types of claims, predictive models can flag cases that might be subject to litigation. Routing such claims through specialist adjusters and streamlining the process can help adjusters reduce litigation costs. As a result, predictive analytics can contribute to reduced fraud costs, reduced loss adjustment expenses, improved adjuster productivity, and reduce the overall claims ratio.  
  2. Demand management. Insurers use a multitude of distribution channels to sell their products, and the selling process takes place over multiple channels, including in person, over the phone, and online. Because insurers face increasing pressure to produce better returns on their marketing investments, they are now using predictive analytics to analyze consumer behavior, which helps calculate their propensity to purchase specific products. Insurers can collect policyholders’ data over time to determine individual policyholder receptiveness to cross-selling other products and when it is appropriate. In doing so, they can see increases in conversion, cross-product, and retention ratios.
  3. Producer acquisition and value management. The average age of an insurance agent is 57. As retirements reduce the number of agents and advisors and economic growth remains low, acquiring, retaining, and enhancing producer productivity has become an even greater priority. Predictive modeling can combine internal insurer data with external socio-demographic data to determine the market potential for specific products. These insights can help the head office sales force improve producer acquisition, retention, and productivity ratios.
  4. Underwriting and pricing. Actuaries and underwriters have typically used predictive modeling to compute risk scores based on things like an individual’s socio-demographics, driving record and behavior, and credit score. They use these predictive scores to determine pricing, as well as automate the underwriting process by setting rates and automatically approving customers for coverage beyond that line. Similarly, they can automatically reject customers who fall below a certain threshold, leaving underwriters to manually evaluate a smaller set of customers. Auto insurers have long used such techniques, but now property, commercial, and life insurers are as well. Predictive modeling plays a critical role in reducing underwriting cycle time, enhancing the ease of business for agents, increasing underwriting consistency, and reducing underwriting costs, all of which lead to better risk pricing. The positive results for insurers include reduced expense ratios, better underwriting results, and enhanced customer and agent satisfaction.

Getting Started

Implementing a predictive analytics program can be daunting, considering the variety of possible applications, the volumes of data that need analysis, and the sophistication of available techniques and tools. Accordingly, a four-stage approach can be employed to achieve results relatively quickly:

  1. Select a business problem and map the decision process. Insurers often undertake large data warehousing or customer data integration initiatives to justify a predictive analytics program. However, it is better to begin by identifying a tangible goal and desired metrics, as well as the critical decisions it has to make to realize them. This top-down approach, which starts by determining the data and insights the company needs and what it has to do to get them, is better than the typical bottom-up approach of agglomerating data in one place before determining what to do with it.
  2. Identify, collect, and analyze data (data mining). Once the insurer has identified its goals and the necessary steps to achieving them, the company should identify the internal and external data sources that can be used to generate necessary insight. 
  3. Build and test the predictive model. Once the company collects and analyzes internal and external data, it can build a predictive model that uses a number of different techniques, including regression, simulation modeling, neural networks, and evolutionary computation.
  4. Institutionalize predictive analytics and an insight-driven culture. By demonstrating the value of predictive modeling in a tangible business application, an insurer can undertake the more difficult task of changing how management and staff make decisions by promoting predictive analytics throughout the company.

Predictive analytics will be increasingly important for insurers, and their claims organizations specifically, who want to streamline their operations and use more insightful data to make faster and better decisions. Insurers that consistently base their acquisition, retention, and management of their customers and agents on predictive models will have an advantage over their peers in targeting, pricing, service, and claims handling.

Anand S. Rao is principal with PwC Diamond Advisory Services.