On 4 November 2026, the Parliamentary Advisory Council for Transport Safety (PACTS) set out a feasibility study into whether a safety rating could help young drivers and their parents choose motor insurance on its safety support as well as its price.[1] The study, the Young Driver Insurance Safety Framework, is funded by Transport Scotland's Road Safety Framework Fund and runs until August 2027.[1] It will review how telematics products measure driving and deliver feedback, along with the evidence linking them to safer driving, before building a prototype consumer rating and testing it with young drivers and parents.[1] Confused.com, Agilysis, Defaqto and Angelica Solutions are contributing insurance and consumer expertise.[1]
The project lands in a debate that has been gathering pace all year. In June, Aviva asked the government to make telematics policies a legal requirement for drivers aged 17 to 24 in their first year, alongside an exemption from Insurance Premium Tax.[2] The Chartered Insurance Institute has opened a consultation on how telematics is used and explained to consumers, following a coroner's report on the deaths of two people in a crash involving a 17-year-old driver, which raised concerns about inconsistencies in how telematics is applied across the market.[3] Department for Transport statistics show that in Great Britain 269 people were killed and 4,466 seriously injured in 2025 in collisions involving car drivers aged 17 to 24.[11]
The Connected Insurance Hub's verdict is that a telematics rating is worth building only if it grades the evidence for each product's safety effect separately from its effect on price. The two are routinely blended in public argument. The best independent evidence on the first is discouraging, and the published evidence on the second cannot separate behaviour from selection. A rating that scores features and feedback design without placing each product's safety claim on the proof ladder would give parents a confident answer to a question nobody has yet tested.
Two mechanisms that are easy to confuse
A young driver telematics policy can reduce an insurer's losses in two distinct ways.
The first is sorting. Drivers who accept monitoring, and who keep their scores high, may already be lower risks. Telematics lets the insurer identify them and price them more accurately, and lets it cancel or reprice the drivers whose data shows high risk. This changes who is insured at what price. It need not change how anyone drives.
The second is behaviour change. Feedback and the incentives attached to it might cause a young driver to drive more safely than the same person would have driven without them. Only this mechanism reduces casualties on the road. It is also the mechanism that a mandate, or a consumer safety rating, implicitly relies on.
The distinction matters because insurer data reflects both at once. Pulling them apart is the job of the Attribution gate, the master gate in the Hub's evidence standard.[4] It asks whether the losses avoided would have happened without the intervention. A lower claims rate among telematics customers cannot answer that on its own, because the customers chose the product and the product then shaped who stayed.
The insurer's figures
Aviva's figures, drawn from its own claims between January 2023 and March 2025, show that newly qualified drivers were more than twice as likely to claim in their first year as drivers with a year or more of experience: 11% against 5%.[2][5] Aviva also reported that new drivers on telematics policies reduced collision claims by nearly 50% over their policy, against about 30% on standard cover.[2]
These figures are insurer-produced. They were published in support of a policy proposal and reported without a published methodology or comparison design.[5] They are treated here as vendor-produced evidence and carry no load in the argument. On the proof ladder, the telematics comparison sits at correlation. Both groups improve as they gain experience. The gap between them could reflect a genuine change in driving, or it could reflect who chose telematics and who was later repriced or cancelled, and the figures cannot distinguish the two. Nothing in that observation suggests the figures are wrong; they simply answer a different question from the one a mandate asks.
The independent trial
The strongest independent test of the behaviour change mechanism in Great Britain is Driver2020, a Department for Transport study run by the Transport Research Laboratory. Its results were published in November 2025.[6] More than 28,000 learner and novice drivers aged 17 to 24 were recruited and randomly allocated to interventions or control groups, with telematics among the interventions tested.[6][7]
The telematics arm used an app that gave feedback on driving style, with modest weekly rewards for good scores and the loss of prize-draw entries for poor ones, designed to mimic a pay-how-you-drive insurance model.[7] In the novice analysis, 2,234 telematics participants and 2,258 controls provided complete data.[7] Allocation to telematics had no detectable effect on self-reported collisions in the first year after the test.[7] Engagement was low: only a minority used the app at all, too few for a reliable collision analysis among those who did.[7] The researchers concluded that offering these interventions voluntarily is unlikely to produce a measurable reduction in collision risk for this age group.[8]
This is credible causal evidence, and on its own terms it points to no effect. Its limits are as important as its result, and the report states them. Collisions were self-reported, not taken from claims, and the sample was smaller than planned, so small effects may have been missed. The follow-up also covered only the first year, in a period disrupted by COVID-19.[7] The trial also tested app feedback with light incentives. It did not test an insurer-priced black box whose premium moves with the driver's score. In the language of the Five Gates, the result has limited Exportability to the products most young drivers actually buy.
Where the two bodies of evidence conflict
The conflict should be stated plainly. Insurer-produced data, which counts as vendor-produced evidence, suggests telematics customers claim less. A randomised trial of voluntary telematics feedback found no reduction in collisions, largely because few people used it. Both can be true. Sorting can lower an insurer's losses while behaviour change remains unproven, and a trial with low engagement cannot show what high engagement would achieve.
What neither body of evidence provides is a test of the intervention now being proposed: compulsory telematics priced through insurance, where use is required by law. The Department for Transport's own 2015 review found that methodological constraints limited what insurance data could show, and concluded that further primary research was needed.[9] Driver2020 followed, and it is the best evidence yet on the voluntary case. The mandatory case remains open.
What a credible rating would grade
For PACTS and its partners, the design choice that matters most is what the rating rewards. A rating built on features, such as the number of behaviours measured or the frequency of feedback, would reward product design whether or not it changes driving. A rating built on evidence would ask each insurer what it can show about its product's effect on collisions, and where that claim sits on the proof ladder.
That second approach is harder, because most products will have little to show. It is also the only version a parent could rely on. An insurer able to present a comparison of matched drivers with and without its product, followed beyond the first renewal, would earn a higher grade than one presenting only the claims rate of its own customers. The Honest Baseline and Persistence gates translate directly into rating criteria.[4] The Hub's earlier work on telematics in commercial lines made a related point: a proposition needs a testable hypothesis linking scores to claims, and a programme that works only for good fleets invites adverse selection.[10]
For insurers and technology providers in the young driver market, the practical step is to start generating that evidence now. Random or staggered allocation of feedback features within an existing book, with outcomes measured in claims, would produce more useful evidence than any amount of customer satisfaction data. It would also prepare a provider for a scalable insurance partnership in a market that may soon be rated on what it can prove.
What the evidence cannot establish
The available evidence cannot show whether mandatory telematics would reduce young driver casualties. Driver2020 tested voluntary use, and insurer data mixes selection with behaviour. No published study found for this article isolates the effect of insurer-priced black box policies on collisions in Great Britain. Whether a rating would change consumer choices is the question PACTS has set out to test, and there is no result yet.[1]
What would change this view
This view would be wrong if an independent evaluation of insurer-priced telematics policies, using a credible comparison group and claims outcomes, showed a reduction in collisions beyond what selection explains. It would also be wrong if the PACTS review found published causal evidence for specific product designs, so that features could stand in for outcomes. Either would justify rating products on design alone.
Technology providers that want to see where their own evidence sits can take the free Readiness Diagnostic. Insurers and providers designing a young driver programme can discuss the opportunity with the Hub.
Sources
- Insurance Edge, "Could a safety rating system help reduce premiums for under 25s?", 4 November 2026 (PACTS Young Driver Insurance Safety Framework announcement). https://insurance-edge.net/2026/11/04/could-a-safety-rating-system-help-reduce-premiums-for-under-25s/ (Commercial-neutral trade press reporting a road safety charity's project)
- Insurance Business UK, "Aviva calls for mandatory telematics insurance for young drivers", 21 June 2026. https://www.insurancebusinessmag.com/uk/news/breaking-news/aviva-calls-for-mandatory-telematics-insurance-for-young-drivers-579635.aspx (Commercial-neutral press reporting insurer-produced data; Aviva figures not load-bearing)
- Insurance Times, "CII launches telematics consultation after coroner raises concerns over young driver safety", 2026. https://www.insurancetimes.co.uk/news/cll-launches-telematics-consultation-after-coroner-raises-concerns-over-young-driver-safety/1458656.article (Commercial-neutral)
- Connected Insurance Hub, "The Five Gates", 5 August 2026. (Internal framework)
- Insurance Times, "Telematics key to reducing high claims rates among new drivers: Aviva", 15 June 2026. https://www.insurancetimes.co.uk/news/telematics-key-to-reducing-high-claims-rates-among-new-drivers-aviva/1458829.article (Commercial-neutral press reporting insurer-produced data)
- Department for Transport, "Driver2020 evaluation", research and analysis, 27 November 2025 (conducted by TRL). https://www.gov.uk/government/publications/driver-2020-evaluation (Primary/independent)
- TRL, "Driver2020 evaluation: summary of findings" (PPR2012). https://www.trl.co.uk/uploads/trl/documents/PPR2012_Driver2020-evaluation-summary-of-findings-d4.pdf (Primary/independent)
- Road Safety GB, "Research finds voluntary interventions unlikely to succeed at improving young/novice driver safety", 8 December 2025. https://roadsafetygb.org.uk/news/research-finds-voluntary-inventions-unlikely-to-succeed-at-improving-young-novice-driver-safety/ (Commercial-neutral summary of primary research)
- Department for Transport, "Telematics research: impact on young and novice driver behaviour", 3 December 2015. https://www.gov.uk/government/publications/telematics-research-impact-on-young-and-novice-driver-behaviour (Primary/independent)
- Connected Insurance Hub, "Beyond the Black Box: When Telematics Becomes an Underwriting Conversation", 28 March 2026. (Internal)
- Department for Transport, "Reported road casualties in Great Britain: younger and older car drivers factsheet, 2025", updated 24 September 2026, Table 3. https://www.gov.uk/government/statistics/reported-road-casualties-great-britain-older-and-younger-driver-factsheets/reported-road-casualties-in-great-britain-younger-and-older-car-drivers-factsheet-2025 (Primary/official statistics)