IE magazine caught up with Neill Slane, Senior Vertical Market Manager, for Claims and Insurance at LexisNexis Risk Solutions, UK & Ireland. We wanted to learn more about how in-depth data can really add useful insights for insurers and brokers alike when it comes to home and commercial property insurance.
IE; Let’s begin with all the changes in home insurance caused by the lockdown, it’s been quite a dramatic few months?
NS; It has, what we are hearing is a huge drop in burglary, flood and fire events. Much of this is anecdotal and we will all have to wait and see what happens as life gets back to normal. From a LexisNexis Risk Solutions perspective our focus is on feeding through the data on buildings or contents claims into something that our customers can use.
There will be a bit of cross-referencing needed to put data into context, so you might ask `what perils claims are linked to subsidence claims, in a given postcode or geo location?’ This will build on the streamlining of perils data at point of quote.
Some risks haven’t changed that much during Covid-19 of course.
IE; Do you think there will be a sharp contrast between residential claims and Commercial?
NS; Although there were a great number of Commercial premises that were locked up and left empty back in March, the interesting thing is how rapidly SMEs have adapted to the changing circumstances. So for example some cafes and pubs have started takeaway service, so they have got people inside the premises, it’s just that the public haven’t been in there as well, so that layer of risk was removed. Other property owners or managers have been going back to check on empty units or shops and there is a certain level of maintenance and systems checking that must be done, even on empty properties.
IE; Tell us more about how the new geo-spatial data package overlays different strands of data, so insurers get a fuller picture.
NS; When we talk to insurance providers we find that they often want extra data that complements the standard perils information they have. What insurers are looking for is data that narrow down the differences between causation, and a correlation, a pattern of some sort. Risk is all about understanding those subtle differences and you can’t really price each risk correctly unless you know all the cross-correlations within your data stack.
For us, it’s very exciting to bring it all together, at sub-second speeds and then the insurer can offer a price online based on that data stream.
IE; What variations can you find within correlating factors affecting property insurance?
NS; Say you do a study on ice cream sales and you find that every time Sunday is sunny, you’re sales are up a great deal. The sunshine is the cause. But then say you see a spike of sales in December, what were the factors that prompted that, and do they correlate in some way to sales spikes at other times of the year? That’s the beauty of overlaying the data and trying to understand patterns that go beyond the obvious.
It’s the same approach with property, both residential and commercial; there are more devices in the home than ever before, so what does the data from those gadgets tell you? Add in data from smart meters, Ring doorbells, wearable tech, connected fridges and Alexa type speakers and you build quite a complex picture, assuming you have all the privacy permissions in place. That’s the exciting thing about the future – there really is so much more potential for personalisation of insurance based around shared data.
IE; So insurers could offer variable insurance, depending on who is working from home, what cars are currently parked on the drive – all kinds of factors?
NS; Yes, the shared data pools let us get to know the customer much more. That teaches the insurer how people behave day-to-day and once we deal with the big challenge of data permissions, then people can get really individual insurance policies in place. The process of ingesting data from various sources, and using it to underwrite policies is maturing right now.
IE; I can see how a video doorbell can feed through visitor info to an insurer and that history then affects the renewal, because the insurer can see how many delivery people visited, if the door was answered, who was in the house and what time of day etc. It really offers tremendous potential to spot crime risk BEFORE a crime is actually reported too – as you can see if someone approaches the front door at 3am or something?
NS; That’s an interesting case yes, and it highlights how much LexisNexis can refine and develop our products based on the demands from the marketplace. If the home owner wants to know who is in their driveway, and they are willing to share that data with their insurer, to get a better deal on Contents, then everyone is learning more and reducing risk. It’s a win for both insurer and consumer.
If you take another step on from that doorbell case and apply it to empty properties on holiday; let’s say the home owner WANTS their insurer to keep an eye on the place while they’re on a long holiday in Australia or NZ for example. The insurer then becomes someone who on the home owner’s side, a trusted guardian. From the insurer perspective, you get to reduce the risk of a claim by getting access to the front door visitor traffic and build a relationship long term.
IE; That kind of thing is great when it comes to renewals time isn’t it?
NS; Here’s an interesting stat; 71% of homeowners stay with their existing insurer after making a claim. So when things are very stressful, and the worst happens, most people are actually reasonably happy with their insurer. Using data to improve that relationship by preventing claims, not just assisting after the event, has to be good news.
IE; Let’s talk floods. How can data history help when a new housing estate is being built, because there obviously isn’t a flood history by postcode and street?
NS; Well you have information from public sources, like maps, building company applications to divert brooks and streams, install new drains, or surface water run-off tech that’s part of the overall project. You know the proximity of water of course and you might well have details on local flood defence works that have been recently carried out, or are at the planning stage.
Nowadays you have a constant scanning of terrain, surveys by drone, aerial photography by public bodies and more, that give you a constant flow of data about almost any given postcode, at least where there is a built environment.
All of these data sources can help insurers predict and model particular scenarios better, whether housing is well established, or it’s a brand new development. In some ways, that’s a good thing because we are now trying to predict events, rather than reacting to them, which is what tended to happen in the past.
IE; In the past motor claims, local vandalism crime reports, or a winter flood were all separate events in terms of reporting. But now much of that can be cross-referenced and that gives insurers a far more detailed history of any given street or area doesn’t it?
NS; Yes, we are out of the silo approach now, information is being shared. Getting platforms to work in sync, in real time, is perhaps the key element in insurance risk now. You can create a good customer experience by using data to properly analyse the risk, price it correctly, and then help prevent claims from happening. When something does go wrong for the policyholder, the data can streamline the settling of that claim much faster. Again, that’s a win from the consumer point of view.
When you can deliver a personal service, based on patterns of data, that delivers on your policy promise, then you’re increasing customer loyalty and winning new customers because you truly understand the levels of risk.
IE; Interesting stuff, thank you Neill.
Be the first to comment