How we rate, rank and price

Stock is ingested directly from FCA-regulated dealer feeds and from our aggregation partner Marketcheck. Prices are pulled at least every 24 hours and on every search; “from £X” figures reflect the lowest active asking price on Motoring Today at page-load time, not historical lows.

Representative monthly figures are real PCP or HP quotes from FCA-regulated brokers, computed against the listing's asking price using broker-supplied APR, term and deposit assumptions. Tap any “£X /mo” chip on a listing for the full representative example. Finance is subject to status.

Compliance is derived from DVLA Vehicle Enquiry Service (VES) emissions data per VIN, then matched against the published TfL ULEZ standard (Euro 6 diesel / Euro 4 petrol) and equivalent rules for Birmingham, Bristol, Sheffield and Scotland's Low Emission Zones. We display the dealer's declared fuel type when VES data is unavailable.

First-time pass rates, common advisories and median MOT life remaining are aggregated from DVSA's public MOT history API, joined to the listing's registration. Statistics are computed across all UK MOTs for that make/model in the trailing 24 months and refreshed weekly.

Annual depreciation is computed from observed list-to-sold deltas in our own transaction history (where available) or modelled from age and mileage against comparable trim. CO₂ and VED bands come from DVLA VES. Monthly finance is the median representative quote across active stock.

Our “most reliable”, “safest”, “cheapest to insure” and “best family” rankings are built from a documented, deterministic model-scores formula computed offline — never by an LLM and never a source figure re-labelled as ours. A model's overall score is a weighted blend of five sub-scores, renormalised over whichever are available (a model needs at least 3 of 5 to publish an “overall” figure, so no headline is built off one or two dimensions): reliability 25%, running cost 25%, safety 18%, family/practicality 17%, insurance 15%. The safety sub-score averages the Euro NCAP star rating (as a percentage) with the four individual pillar percentages — adult occupant, child occupant, pedestrian and safety assist — wherever Euro NCAP publishes them for a model, then applies an era discount so an older, less demanding test protocol can't outrank a modern one at the same nominal star count: full credit for tests from 2020 onward, tapering to an ~18% reduction for tests at 2010 or earlier, linear in between. Star ratings and pillar scores are independent Euro NCAP assessment data — Euro NCAP is not the manufacturer — and our score is Motoring Today's own transform of that data, not a Euro NCAP figure itself; where Euro NCAP's test year for a model is old enough to be tested under a materially different protocol, the ranking article discloses it rather than treating all star ratings as equivalent. Reliability prefers DVSA's national MOT dataset (the age-controlled first-MOT pass rate) where a model has a large enough national sample, and falls back to our own marketplace MOT sample otherwise — every ranked table marks which source backs each row, rather than labelling marketplace-derived rows “National DVSA”. The “typical UK price” shown alongside these rankings is a MarketCheck market-wide median asking price — there is no national used-car price source — bound to the same vehicle generation as the cited rating wherever we can confirm one (shown with a listing count and snapshot date); where we can't confirm a generation match, the table and the accompanying price note say so rather than implying one.

Our market reports and price indices are computed from live UK used-car listings aggregated by Marketcheck, our data partner — analysis and findings are Motoring Today's own. Figures are drawn from Marketcheck's UK coverage (currently ~550,000 live listings): a large, representative sample of the dealer market, but not the entire UK car parc, and skewed toward advertised dealer stock rather than private sales. Medians and days-on-market are recomputed monthly so month-on-month change reflects real movement. Regional comparisons (e.g. prices by city) are drawn from that city's live stock; where we compare cities we note that a raw median gap can partly reflect a different mix of ages, models and trims, not just price — and we age/mileage-adjust or flag it where it matters.

We don't include private-seller listings or auction stock. We don't track historical price changes older than 90 days. Where a dealer hasn't provided a VIN we fall back to model-level averages — flagged on the listing.

Daily. The valuation model re-runs against fresh live-listing and market data every 24 hours, so a rating reflects the market as it is today — not last quarter.

No. Ratings are computed automatically from price and market data, and dealers have no input into the model. If a price changes, the rating changes — nothing else moves it.