By Jeffrey Skelton, MD, LexisNexis Risk Solutions, Insurance, UK&I
The power of data enrichment to understand risk in the insurance market is growing. This is due in large part to data sharing by the insurance industry via contributory databases so that decisions can be made on the wider market’s experience with that customer. This is not dissimilar to the lending sector’s use of market data. In the insurance sector, analysis of this data is revealing new insights to help the market price and underwrite with a clearer view of the potential loss from claims or cancellations.
In addition to contributory data on policy history, intelligence on quote behaviour, along with property characteristics and geo-spatial data around environmental risks is now at an insurance provider’s fingertips.
At the same, the emergence of the Internet of Things (IoT) is creating data flows from cars, homes, offices and personal devices. In motor and home, links are being created from these IoT assets to the insurance market, opening the opportunity to mitigate risk in real-time.
Data must be actionable
There is no shortage of data the insurance market can now call upon, the key is extracting insights from data that is meaningful and actionable then using it at point of quote and throughout the customer journey.
Insurance quotes and claims in seconds
With data as the backbone of decisions, a future in which an insurance application can be completed in less than a minute or a claim settled in the same time, on a smartphone is not that far away. Already a life insurance quote in the US has reduced from 30+days to minutes through the data modelling used to support decisions at point of quote.
Motor insurance transformation
The UK car insurance market is undergoing the biggest transformation through data enrichment. Over 80% of the car insurance market are sharing policy history data including No Claims Discount information to immediately validate No Claims Discount bonus’ (NCD) at point of quote. The value of policy history data goes beyond operational savings and can help address application fraud, understand claims risk and the possibility of future cancellations. For example, when we ran a piece of analysis[i] we found:
- Past cancellations can equate to 70% higher loss cost, a person with two prior cancellations are more than twice as likely to cancel again.
- An individual with more than one NCD entitlement at any one time has a 33% higher loss cost and those who have had a NCD downgraded in the past are 60% more likely to cancel.
- The more often people switch vehicles the more likely they are to cancel.
- There is a 50% higher loss cost when a customer has previously had a gap in cover and one gap in cover in the last 5 years means they are 55% more likely to cancel.
Switching linked to claims and cancellations
Knowing that motor insurance is the most ‘switched’ of all insurance products and retention is a business objective for insurance providers, it is also possible to look at shopping behaviour to see how often and at what stage the customer switches and how that correlates to claims and cancellations. Our analysis has shown that when a quote is obtained for same day cover there is a predicted 32% higher loss cost. Furthermore, there is a 91% higher chance of early cancellation.
Knowing more about named drivers
Within the insurance market, the understanding the risk of named drivers has been a challenge – given 42% of policies in our database have a named driver present this is a serious gap in knowledge. Again, policy history provides a greater understanding of risk – our data science team has already uncovered that named drivers on cover that had two prior cancellations lead to 40% higher loss cost and are also twice as likely to cancel in the future.
Contributory data on claims
In order to help insurers better understand loss costs and drive them down so that consumers across the industry can benefit from lower prices, the next phase for us is to bring in claims data contributed by the market. This is starting in home, followed closely by motor and in the future commercial segments. Bringing in greater insights about claims – the circumstances, the settlement figure – will add a further layer of granularity when assessing risk.
To date, underwriting tends to be siloed by line of business. But with greater data insights the opportunity opens up to apply data used for risk assessment in one product line to help price and underwrite another. For example, an insurance provider assessing the flood risk for a property and the crime level, could apply that intelligence to underwriting a car insurance policy. Real-time data on flood events is a great example of an opportunity to mitigate the risk for both the property and the vehicle.
Cohesive view of the customer
With a more cohesive view of the customer the opportunity to segment and provide tailored services opens up to help move away from price-driven decisions. Insurance can start to help mitigate risk rather than step in only when disaster strikes. Connectivity from assets further enhances that potential and, again, motor is where we are seeing that transformation first.
Connected car insights
The framework for connected cars to communicate directly with insurance companies is already in place bringing in a completely new level of data enrichment. Connected car data has the potential to tell insurers the car’s mileage which will be valuable as travel behaviours may alter, post-pandemic. It can also confirm the car’s location, how it is driven and what ADAS features are present and being activated. It offers the opportunity to deliver value added services to customers and will create new risk modelling opportunities for the whole of the sector. There is acceptance that in time, the proxies for risk currently used by the market will reduce in importance – the car will become the lead factor in determining risk.
Connectivity in our homes and buildings also holds the opportunity to change the market dramatically through the data insights it could deliver. While reports suggest the pandemic caused a spike in investment in home technology[ii] and video gaming[iii], smart home gadgets are still relatively expensive, and the reason for purchase is more to do with convenience, a burglary or a previous escape of water claim than for a discount. As such, change is happening at a much slower pace. However, if you consider developments such as smart home cameras and door sensors to reduce burglary claims, there remains much potential for the market to link with technology suppliers to encourage adoption. Only through wider adoption can insurance sector really start to understand the impact of smart home tech on risk.
Flags for Fraud
ID verification and validation at point of quote is a key component of the data enrichment process. But as is the case in banking, there is a need to create balance between providing a frictionless consumer experience with fighting fraud and financial crime.
The sector can already take advantage of combined powerful analytical technology and expansive data assets to help detect and prevent fraud with proven identity verification and identity authentication solutions.
In addition, fraud flags associated with named driver risks and cancellations put insurance providers on alert while data from connected things can also do much to help tackle fraud. Telematics data from usage based motor insurance for example can already help validate the circumstances of claims.
Digital footprints linked to past incidences of fraud are also being used as a solution to put insurance providers on guard at quote and claim.
There is little doubt that the demand for data enrichment is only going to grow as insurance providers respond to the changing needs of their customers, and data from assets will be a big part of this. As such, consent management will be a key focus as part and parcel of the process of building products and services to ensure transparency for consumers and their understanding of value in sharing this connected data.
[i]This article contains results of analysis carried out by LexisNexis Risk Solutions UK Limited on available data within the LexisNexis® Motor Policy History database. The analysis was completed within a fixed period and does not purport to represent the results of any identifiable customers. The statistical analysis reported is provided “as is”, nothing arising from the data should be taken to constitute the advice or recommendation of LexisNexis Risk Solutions.