Articles - Power and Pitfalls in Using External Data

 By Dan Marshall, Managing Director, UK Insurance, LexisNexis Risk Solutions
  

  Amid increased competition from aggregators and brokers in personal lines, general insurers face new pricing and underwriting challenges as well as opportunities presented by technological developments. Increasing adoption of online channels puts pressure on insurers to verify online applications quickly and accurately. Meanwhile, advances in the capture and storage of data, from telematics to social media, have made available more information about consumers than ever before.
  

 In response, many insurers are looking more closely at the smart use of external data sources at the time of quote to enable more competitive pricing, increase profitability and improve risk management. However, insurers must be sure to select and manage their data sources with care, being mindful of privacy, data quality, scope, responsiveness and searchability.

  

 Current trends

 Typically, insurers are applying external data with two main goals: to enhance their pricing and underwriting functions; and to improve the customer experience. Three main trends are at play:

 
 • The increasing need to verify applicant or risk details online. Reliance on consumers to self-report at the time of price quotation has led to significant under-reporting of true risk, costing insurers more than £1.9 billion* each year in undetected fraud alone. Without access to accurate and reliable sources of data to verify applicant information, insurers are increasingly exposed.
 • Intensified price competition from aggregators and brokers. Insurers are facing finer margins and reduced profitability due in part to consumer price sensitivity. In addition, the immediacy of online quoting creates a more rapid impact of pricing mistakes and less room for error.
 • The availability of more and new sources of data. No longer limited to public credit records, insurers can now take advantage of other sources of data. Innovation in technology has led to tools that can track driving behaviour (telematics), as well those that can correlate a customer’s shopping behaviour online and in-store. Consumers using social media create a digital footprint that also offers insight into their behaviour.
  

  

 External data sources can improve the accuracy of pricing and underwriting by:

 
 • Refining risk predictions based on records of claims frequency and severity
 • Feeding data to price elasticity models
 • Determining a consumer’s propensity for add-ons and related products.
  

 Furthermore, as insurance is increasingly commoditised, many insurers seek to create a competitive advantage by improving the customer experience. At the point of quote on direct channels, they can use external data sources to pre-fill fields. Not only does this streamline the process, but also improves the application’s accuracy by reducing entry errors and verifying, rather than collecting, information.

 
 Applications of external data

 
 Direct motor insurers have been using external data for some time, such as using vehicle look-up and public credit data at the time of quote, and CUE checks at time of sale. They also use postcode-based data on perils and socio-economic profiles. More recently, some insurers have incorporated external data into their processes through arrangements with software houses and broker platforms.

 
 However, in the United States, insurers use external data more extensively—notably, to enable automated decisions at the time of quote and sale. Common data sources include past claims and policy history, motor vehicle and driver licensing information, public records and credit repayment behaviour. Within the next year, similar data sources are likely to be available in the United Kingdom, including DVLA driver data, policy and claims data, and even private credit data.

 
 Additionally, the technology of capturing and storing data has changed dramatically, resulting in access to a number of new data sources. Consider:

 
 • Shopping behaviour data, including supermarkets and online channels
 • Previous insurance quote and purchase data, which can be indicators of ‘gaming’ (i.e. changing age or postcode to get a better rate) and price elasticity
 • Telematics data, which can enable more accurate assessment of driving risk
 • Social media, where profiling friends and mining status updates can help determine risk and monitor the customer experience.

 
 In navigating this data-rich landscape, insurers must consider a variety of issues to be able to effectively select from and manage sources of external data.

  

 Five consideration factors in incorporating external data

 
 As insurers incorporate more external data into their business, five factors are important to consider: privacy, quality, scope, responsiveness and searchability.

 
 Privacy
 Given the scope and nature of the external data that is useful to insurers, protecting consumer privacy is imperative. Has the consumer given consent for the collection and use of their data for the stated purpose? Is the data being held securely? Is it being used for approved purposes?
 Privacy extends beyond compliance with the Data Protection Act. If insurers are to fully leverage the power of external data, it behoves them to demonstrate to the public that the industry can manage and use the data appropriately—and that doing so can result in lower prices for the majority of consumers.

 
 Quality
 External data must be accurate—not only for utility, but also to win the public’s confidence. The best data providers will rigorously audit the data for missing or inaccurate fields, and where they source the raw input from other parties, create incentives and penalties for corrective action to ensure data quality. Furthermore, the best data providers will allow consumers to have access to data held about them, as well as the ability to correct any errors.

 
 Scope
 With the increased availability of external data, insurers will want to design new pricing models, test different sources, and develop new pricing or underwriting rules. However, doing so can strain existing analytical as well as IT resources, which must maintain connections to the different data sources and manage the enquiry process.
  

 One way to reduce the resource impact and reap maximum benefits is to work with a data aggregator. The best data aggregators can be powerful analytical partners. For instance, they can:
 • Help insurers conduct retrospective analyses to validate sources.
 • Build multi-source scoring models to improve pricing and underwriting decision making.
  

 In addition, data aggregators can be productive IT partners. They can:
 • Negotiate to source and assemble a variety of data sources.
 • Set up and maintain a single link between data sources and an insurer’s system, particularly pricing tables and underwriting systems.
 • Assist in reformatting data contributions and enquiries.
 • Reduce the IT changes required of an insurer by hosting parts of the pricing and underwriting decision systems.
  

 By working with a data aggregator, insurers can incorporate external data while allowing their typically scarce analytical and IT resources to focus on adding value in other high priority areas.

 
 Responsiveness
 If insurers are to incorporate external data at time of quote and time of sale, they must have timely and reliable access to the data. As comparison shopping becomes more prevalent, quote volumes will continue to increase. The best data providers will be able to manage high volumes, provide rapid response times and minimize downtime—an increasingly important factor for a business that is ‘always on’.

 
 Searchability
 Even if an insurer qualifies external data sources with all of the previous considerations, there is little benefit if they cannot search the database and match a high proportion of applicants to accurate records. The best data providers leverage sophisticated linking and matching algorithms to enable high and accurate match rates, based on relatively few characteristics, such as name, address or date of birth.

 
 Next steps

 
 External data can help insurers better understand the risk that they are underwriting. With a complete picture of a driver and vehicle, they can better assess risk and detect fraud. And by feeding external data into analytical models, they can quote more accurately and attract desirable risk profiles at the right price point.

 
 The benefits trickle down to consumers, the vast majority of whom will receive lower, more accurate pricing. Further, external data can enable an insurer to enhance the customer experience throughout the policy lifecycle—from streamlining at the time of quote to processing claims more quickly and refusing fewer claims as a result of inadvertent misrepresentation. In a market where insurance is increasingly commoditised and customer opinion is easily disseminated across the web, the ability to attract and retain customers is critical. A business that uses accurate, reliable data in an appropriate manner will minimise ‘pain points’ for consumers and will create a strong brand which consumers will value.

 
 In future, the discussion will shift from protecting proprietary/internal data sources to how best to use external data. Innovative insurers will use intelligent combinations of internal/external data and analytics to drive new insight, ultimately creating a competitive advantage.

 
 Dan Marshall is Managing Director, UK Insurance, LexisNexis® Risk Solutions. Dan has managed a variety of data businesses for LexisNexis in the UK and Europe, and is well versed in the challenges of making complex data solutions work for real-time transactional decision making. Dan was appointed to lead the development of LexisNexis’ UK Insurance business in May 2011.