By Selim Cavanagh, Vice President, Insurance, UK and Ireland, at LexisNexis Risk Solutions
In fact some recent research (1) shows that on average, only 16% of motor insurance customers did not shop around at renewal over the past 12 months.
They say if you want loyalty, get a dog. But some of the UK’s best loved brands brush aside this cynicism year after year because they deliver value and more importantly, make their customers feel valued.
The new requirement for insurers to publish last year’s insurance premium with the following year’s renewal quote has almost certainly exacerbated this switching behaviour. Bearing in mind it can take up to six months for an insurer to recover the costs of on-boarding a new customer, there is a clear and urgent need for change.
While every insurer has their own pricing strategy, the best pricing decisions are made on the most accurate, single view of the customer and achieving this single, rounded view, at speed, has been one of the biggest challenges for insurers, given the increased volume of switching activity and the vast swathes of customer data now available to the sector.
New risk attributes from external data sources have been aiding this process by enabling insurers to summarise a range of attributes into a single score in much the same way credit scoring in the consumer finance market has worked. In fact many insurers are benefiting from data sharing and rely on this data to get a more accurate view of the customer. However, the risk insights gained by linking and matching customer datasets across the whole insurance business – for example, understanding that a new motor customer was once a home insurance customer – is an area that is not currently widely factored into risk scoring.
For smaller insurers and intermediated business via brokers, without access to the mass of customer data and claims history of larger operators, standard scoring solutions based on credit scores have provided the summarised risk indicator they need to place business. Again these have their limits as they are not based on specific insurance attributes such as claims history, policy history or other contributory data assets, having the capability to factor in risk information from linking and matching customer records.
The next generation of scoring solutions for the insurance sector must tackle these issues head on. Insurance scores need to be built on specific insurance attributes with data directly related to how insurers see performance in the real world.
At the crux of this industry specific score is ensuring that all prior knowledge, links, contact with the customer is consolidated into that one view and one score using data matching and linking technology. The more information you know about aspects of a customer’s life such as their address history, prior policies held across all lines of business, whether they have made a claim on their household insurance for example, all contribute to a better understanding of what they may want, how likely they are to renew and the price and product the insurer should provide. We know from our tests that when you can bring multiple customer data sets together, and match these to create that single view, underwriting performance improves by a wide margin (2.)
• Motorists with a gap in cover in the prior year have a 50% higher loss cost that those that don’t. Motorists with a gap in the last five years have a 30% higher loss cost.
• Motorists with a policy cancelled prematurely have a 33% increase to loss cost, where having two or more has a 70% increase to loss cost.
Historical gaps and cancellations are new and unique insights into a person’s past motor insurance behaviour that have been found to not-correlate with other data sets already used in insurance. Therefore they add incremental value, over and above existing data enrichment sources.
Building in claims data to the score would be another key differentiator from any other standard scoring solution. Align this information with additional factors such as current policies and no claims discounts, cancellation history, gaps in cover, NCD misrepresentations, purchase timing and switching behaviour. Also Public Records data such as CCJs, council tax banks, directors of UK businesses and insolvencies and you will have a score that is truly insurance specific.
With data matching and industry-wide claims history at its core, a true industry score built on attributes relevant to insurance could be highly predictive of insurance loss on new business. By building the score from data from across the sector, the ability to see patterns in behaviour or the customer’s lifestyle will provide a much more rounded view of risk. The deeper use of such data enrichment will increasingly become a major competitive force for automation and delivering a better customer service – the factors that the consumer is looking for, in return for giving their loyalty.
Motor insurers and brokers are constantly innovating to address the challenges of the market. The pricing issues in the sector need to be fixed, that is widely recognised. A score based on real life insurance behaviours and data attributes could not only help the sector price much more effectively but allow them to become more sophisticated in the way they segment and engage with their customers to engender greater loyalty levels. With the single customer view, the ability to understand the customer’s wants and needs increases immeasurably, helping to boost retention and reduce the chances of overpricing.
(1) Consumer Intelligence Insurance Behaviour Tracker https://www.consumerintelligence.com/articles/the-tipping-point-for-shopping-around
(2) Analysis of policyholder data held on the LexisNexis Risk Solutions Policy History Motor Database
|
