Insurers test the limits of telematics’ big data
We’ll send you a myFT Daily Digest email rounding up the latest Big Data news every morning.
The information that insurers receive from telematics devices is revolutionising the way car coverage is priced. Providers are finding more imaginative ways of using the data.
Telematics — or the use of black boxes installed in vehicles to monitor, measure and assess what is actually happening on the road — has been in use for several years and is growing fast.
According to Ptolemus, a specialist consultancy, there were 12m telematics-based insurance policies around the world at the end of last year, and the company expects the number to rise to 93m by 2020, equivalent to 10 per cent of the global motor market. By 2030, says Ptolemus, half the world’s vehicles will be insured with telematics policies, generating €250bn of premiums for the industry.
The two biggest users of the technology are the US and Italy, each with about 4m telematics policies, but they have adopted it for very different reasons.
In Italy, crime was the major driver at the start. “Originally it was used to enable owners to find stolen vehicles,” says Frederic Bruneteau, managing director of Ptolemus. “And fraud detection has now been added as an objective. The fact that the technology can record a crash in a reliable manner is key.”
In the US, the aims are different. “It is purely focused on pricing and upfront discounts,” says Jonathan Hewett, chief marketing officer at technology provider Octo Telematics. Initially insurers would leave the boxes in customers’ cars for just a few months in order to assess driving style, with a view to giving discounts to safer drivers. More recently, however, they have taken to leaving the technology installed permanently.
The UK, the third-biggest market for telematics, has taken a similar approach to the US. “In the UK it is mostly targeted towards young drivers and fleet vehicles,” says Mr Bruneteau. “The challenge for insurers is to go beyond that.”
Ambitions to widen the customer base in each market will depend partly on the ability of the insurers and the hardware providers to cut costs. If a driver’s premium is $500 per year, for example, it might be worthwhile for the insurer to spend money buying and installing equipment. On cheaper policies it is harder to make the economics work.
But costs are falling. “The whole [telematics] industry has taken on the challenge the insurers have set to bring down the cost of the technology,” says Mr Hewett, adding that the technology is getting “more sophisticated, smaller and cheaper”. Many insurers, such as Carrot in the UK, now use smartphones rather than black boxes to analyse their customers’ driving.
Added sophistication is enabling telematics to have uses well beyond analysing driving style. Increasingly, insurers are using data provided by the devices for other purposes.
Reacting to accidents is one example. Charlotte Halkett, marketing actuary at UK based telematics insurer Insure The Box, says the company analyses data coming from cars to spot when incidents might have occurred. “If the G force passes a certain level, it sets off an alarm in our call-centre. If the car is [then] stationary, we proactively try to contact the customer, and often we get a breakdown truck out. About 19 times a month, we call out the emergency services based on data about the G force, the time of day and the location,” she says.
Fraud prevention is also growing in the UK. Insure The Box recently used the data provided by telematics equipment to uncover a fraud ring involving 31 claims from seven accidents over five months that could have cost the company around £500,000 in total.
Expected global value of telematics policy premiums in 2030
Despite these multiple uses, telematics has not taken off everywhere. According to Ptolemus, reasons for that include concerns over privacy, cost, the time it takes insurers to set up telematics-based products and problems convincing brokers and agents to sell telematics policies to their customers.
Over the longer term, telematics faces a potentially more serious threat. The other big technological development that affects motor insurance is the growing use of autonomous features in cars and, eventually, the introduction of driverless cars. Technology that analyses driver behaviour, say some analysts, will be of little use when humans are no longer driving the cars.
Mr Hewett is not worried by these developments. “There is going to be a very long parallel period when autonomous vehicles will be on the same roads as non-autonomous vehicles, pedestrians and cyclists,” he says. The ability to analyse how humans drive cars, in other words, will be important for some time to come.
Get alerts on Big Data when a new story is published