What’s the value of having in-house data capability? Perhaps the best business case for the function is made by the quantity and quality of data-driven insights and forecasts made possible by having that in-house capacity to gather, collate and disseminate information – and in doing so identify gaps, trends and opportunities across the insurance ecosystem.
Percayso Inform’s move to bring its data functionality in-house is a case in point noted Kieran Fisher (pictured) who moved across to join the data intelligence provider and support the development of its vehicle data insight proposition last year. Percayso recognised the benefit that being data-agnostic while still able to partner with data providers would bring its entire value chain, he said, and he’s relishing the opportunity at hand to work with players across the market to generate unique vehicle insight data.
A key area of focus for Fisher and his team has been uncovering the ‘must-have’ vehicle data variable trends that will be shaping the motor insurance space over the next 12 months. This data has been gathered from discussions with the key decision makers across the piece - the analysts, underwriters etc. at the top 10 UK insurance companies – and working out their pain points, their areas of focus and what they need in terms of unique insights.
From these conversations, four standout themes have emerged:
ADAS (or advanced driver assistance systems) technologies - from optional extras, to safety-critical features, to driver aids - have been on the tip of everyone’s tongue for years now. But while they’re not necessarily new to the industry, he said, exciting evolutions have occurred that could propel them into more general usage.
Fisher highlighted that traditionally ADAS technology faced two main concerns – a lack of valuable data and the lack of standardisation of the language and terminology used around this tech.
“On the first point, when analytics were done on the data, for your average private car motor insurer, there wasn’t enough of the book with decent ADAS tech to make the investment worthwhile,” he said. “But actually, I think that with [certain] standard-fit driver aids becoming an industry mandatory item from 2022… combined with the fact that things that used to be reserved for the premium makes – your BMWs and Mercedes - are now starting to creep their way into your run of the mill, everyday cars – your Peugeots and Fords - that’s only going to increase.”
As the ADAS coverage ramps up, the benefits of understanding the technology around this functionality will become more visible. And with the way the industry is going and with manufacturers now using ADAS tech as standard in new cars, Fisher believes the time is right for insurers to get started on that journey.
“On the second side of that, you have [so many] mainstream manufacturers, each based in a different country and speaking a different language, each with their own versions of the technology,” he said. “People were finding it interesting, but incredibly hard to consume, it was just too raw. So, we’ve basically derived a way to standardise the ADAS tech that’s fitted to vehicles across manufacturers.
“It’s quite a manual process but we have been able to look at manufacturers’ standard terminology, decode it and map it into something that’s a bit more meaningful. Not only does it make the entire process so much simpler… but it also means it’s consumable at the point of quote. That data is available, consumable and understandable by machine learning algorithms in 0.2 seconds.”
When Fisher first joined the data intelligence space, he was amazed to find that vehicle valuations did not operate under a standardised language. Understanding the value of the asset you’re insuring is critical, he said, and during his time in the industry, he has come to recognise that accuracy is key. Because ultimately, there’s no point in paying for a valuation model that is only as accurate as the customer-declared vehicle valuation.
“So, we’ve derived a really interesting live, AI-driven valuation model,” he said. “And the AI is the really interesting piece there because of course, tracking vehicle valuations in real-time across the industry is already incredibly high-tech. But by using the AI model, we can consider over 100 different vehicle parameters which gives us whole-of-market coverage. At last count, our valuation coverage of vehicles on the road was something like 98.5%.”
Percayso’s approach to utilising the data from previous adverts is pretty unique, Fisher said, and is made possible by a crawler that looks at the market every day and gathers intel on the vehicle in question. With a database of some 650 million previous adverts, Percayso has access to a wealth of information – from images, to pricing, to mileage, to seller type, to its location.
“All of this data is quite static data which can be consumed quite easily but I think actually the most impressive part of the data within a previous advert is the actual ad text,” he said. “Because, when you’re selling a car, you shout about its features, its functions, its history, its maintenance, even – if you want to be honest – any pre-existing damage or modifications.
“That raw text block is hugely rich for pricing, risk and fraud analysis. So, we’ve derived a tool that looks for trends and keywords within that text… And [the information we derive], despite the fact it’s coming from a raw text field, we can deliver at the point-of-quote in sub-two second response speeds to throw those flags into the mix from a pricing, risk and fraud perspective.”
If you’re looking for valuable, timely raw vehicle data there are surely few sources better than an MOT? Percayso obtains this data directly from the DVLA, Fisher said, which ultimately is a live system but what’s interesting to see is that while some insurers are already consuming the most basic MOT data, few are digging deeper into the wealth of insight it can yield.
To his mind, the real benefit is around understanding the condition of the vehicle. When insuring an asset, he said, understanding that asset as much as possible and how likely it is to be involved in a claim is what insurers should be doing. So, Percayso is taking it one step further and starting to examine the results of MOTs, looking for advisories and failures and categorising these into different tranches.
This enables Percayso to deliver an MOT ‘risk score’ based on the condition of the car and, crucially, its maintenance at the point-of-quote and at speed, he said. And be able to understand these trends on a vehicle-specific basis and deliver these insights in a meaningful, consumable way is an interesting proposition that is translating well across the insurance marketplace.
Fisher and the team are seeing significant appetite from insurers to engage with new data variables, which he highlighted is unsurprising given that everybody is looking for the newest exciting data to give them an edge in today’s competitive market. Having access to these processed results on vehicle data is a key way for companies to stand out but it also allows them to price better and write better business.
“There’s a big focus at the moment on only writing desirable customers, only writing customers where you’re happy on the risk - quality rather than quantity,” he said. “That’s the message I’m getting across the board at the moment. So being able to understand that and being able to get as much data as possible so you can derive who is a low-risk customer is what’s going to make businesses profitable in 2023.”
What trends do you see governing the motor insurance market in 2023?
