Compare The Recall Records Of Up To Three Car Models Side By Side, Using DVSA Recalls And DfT Licensing Data
Pick up to three models to compare their UK car recall records since 2020: how many recalls, how serious, which systems and how many cars each affected, alongside how many of each model are on the road. Then check the actual car's registration, because one car's open recall matters more than any model's total.
| Mercedes-Benz | 421 | 18.04 | 833 | 0 |
| Ford | 208 | 3.43 | 378 | 1 |
| BMW | 143 | 6.69 | 275 | 1 |
| Peugeot | 138 | 9.16 | 421 | 1 |
| Vauxhall | 135 | 3.34 | 300 | 3 |
| Citroën | 131 | 12.25 | 414 | 4 |
| Toyota | 100 | 3.19 | 250 | 0 |
| Volkswagen | 88 | 2.11 | 220 | 3 |
| Audi | 76 | 3.63 | 158 | 3 |
| Land Rover | 66 | 6.89 | 159 | 0 |
| Fiat | 58 | 4.48 | 199 | 0 |
| Hyundai | 58 | 4.77 | 102 | 0 |
| Porsche | 57 | 22.94 | 122 | 0 |
| Kia | 51 | 3.81 | 87 | 0 |
| Nissan | 45 | 1.71 | 162 | 1 |
| Jeep | 42 | 44.43 | 112 | 0 |
| Volvo | 35 | 4.47 | 175 | 1 |
| Renault | 33 | 1.62 | 196 | 1 |
| Jaguar | 31 | 8.67 | 104 | 0 |
| DS | 31 | 62.34 | 32 | 0 |
| RAM | 28 | n/a | 28 | 0 |
| Mazda | 27 | 4.6 | 103 | 1 |
| SEAT | 27 | 4.2 | 72 | 1 |
| Suzuki | 20 | 4.13 | 90 | 1 |
| Škoda | 19 | n/a | 61 | 0 |
| MINI | 19 | 2.06 | 21 | 0 |
| Bentley | 18 | 66.18 | 46 | 0 |
| Tesla | 18 | 5.93 | 26 | 0 |
| Alfa Romeo | 17 | 24.63 | 55 | 0 |
| Dodge | 17 | n/a | 24 | 0 |
| Lamborghini | 17 | n/a | 22 | 0 |
| Honda | 16 | 1.74 | 96 | 0 |
| Maserati | 15 | n/a | 48 | 0 |
| Subaru | 13 | 22.61 | 37 | 0 |
| Lexus | 12 | 5.65 | 42 | 0 |
| Polestar | 12 | 20.82 | 12 | 0 |
| Ferrari | 10 | n/a | 48 | 0 |
| Dacia | 10 | 3.23 | 18 | 1 |
| Chrysler | 9 | n/a | 71 | 0 |
| McLaren | 9 | n/a | 11 | 0 |
| Genesis | 9 | n/a | 9 | 0 |
| Aston Martin | 8 | 27.24 | 35 | 0 |
| Rolls-Royce | 8 | n/a | 21 | 0 |
| Mitsubishi | 7 | 4.16 | 112 | 0 |
| MG | 7 | 1.48 | 12 | 0 |
| BYD | 5 | 5.23 | 7 | 0 |
| Chevrolet | 4 | 9.99 | 35 | 0 |
| Lotus | 3 | n/a | 17 | 0 |
| Isuzu | 3 | n/a | 10 | 0 |
| Infiniti | 2 | n/a | 15 | 0 |
| LEVC | 2 | n/a | 15 | 1 |
| SsangYong (KGM) | 2 | n/a | 10 | 0 |
| Caterham | 0 | n/a | 8 | 0 |
Tap a column heading to sort. Per 100,000 cars uses DfT licensing figures for 2026 Q2; n/a means too few cars licensed to compare.
Recall counts show faults manufacturers have agreed to fix, not how often cars break down. A brand that recalls readily can look worse than one that does not.
How To Read The Comparison
Read the recalls per 100,000 licensed cars first, then the stop-drive count and the systems affected. The raw number of recalls is the least useful figure on its own.
- Recalls since 2020
- DVSA recall references launched since 1 January 2020.
- Per 100,000 cars
- Car recalls since 2020 divided by the make’s cars licensed in the UK, times 100,000.
- Stop-drive
- Recalls whose notice tells owners not to drive.
| Metric | What it tells you | What it doesn't |
|---|---|---|
| Recalls per 100,000 licensed cars | How often the model is recalled relative to how many are on UK roads1 | How often cars actually fail |
| Stop-drive recalls | Whether any recall told owners not to use the car | Whether those recalls have been fixed on a given car |
| Systems affected | Which parts recur: brakes, airbags, fuel, software and so on | The severity of each individual case |
| Biggest single recall | The scale of the largest campaign, in vehicle-records | How many cars are still unrepaired |
| Share fixed by software | How many fixes are quick updates rather than new parts | Whether an update has been applied to your car |
| Months from build to recall | How long problems took to surface | When the next recall will arrive |
| Still licensed | How many of the model remain on UK roads2 | The condition of any one car |
Models with fewer than five recalls since 2020 show a list of those recalls instead of percentages, because a rate built on one or two recalls would mislead.1
Does A Recall Mean A Car Is Unreliable?
No. A recall means the manufacturer found a safety defect and agreed a free fix under DVSA's oversight. Reliability surveys measure breakdowns and faults, which are usually not recalls.3
The two measure different things. A recall is a planned repair for a defect that could affect safety, organised by the manufacturer and often done before most owners notice a problem. A reliability survey counts what goes wrong in everyday use: electrical gremlins, worn parts, failed sensors. A car can have several recalls and still be dependable day to day, and a car with no recalls can still be costly to run.
What matters most is whether the recalls on a given car have been done. DVSA's last published figures show 89 % to 98 % of cars and light vans were fixed by the time recalls closed.4 A commercial study, whose method is not public, has claimed that a large share of recalled UK cars remain unrepaired.5 Either way, the specific car's open-recall status matters more than its model's count. See what a car recall means.
Why per-100k Figures Are Fairer
Best-sellers collect more recalls simply because there are more of them on the road. Dividing by the number licensed, from DfT figures, shows the rate instead of the size.1
The same logic works at brand level. Since 2020, Ford has launched 207 UK recalls and Mercedes-Benz 421. Ford has about 3,703,510 cars licensed in the UK against 1,768,348 for Mercedes-Benz, so per 100,000 cars the gap is much wider: about 3.43 for Ford against 18.04 for Mercedes-Benz. Models work the same way. A rare model with two recalls can have a higher rate than a best-seller with ten. See which car makes have the most recalls.
Shared Platforms, Shared Recalls
Models built on the same platform or with the same engine often share recalls, so switching badge does not always avoid a campaign.6
Common examples include:
- Stellantis 1.2 PureTech engines in Peugeot, Citroën, Vauxhall and DS models, recalled together for belt and fire-risk problems. See PureTech recalls.
- Jaguar and Land Rover models that share components and sometimes share recall codes.
- Volkswagen Group models across Audi, SEAT, Cupra, Škoda and Volkswagen that use the same parts.
Shared recalls are also why two models from different brands can show almost identical records. The comparison flags recalls that cover more than one of your chosen models, so you can see where a problem is shared rather than unique.
Before you buy
Check The Actual Car, Not Just The Model
Run the registration through the checker and ask the seller for proof that open recalls were done. A dealer must fix safety recalls before selling to a consumer.7
Which Car Is It?
Choose the make and model so we can match recalls to your car.
A private seller does not have the same duty, so the check matters even more when buying privately.7 Ask for the dealer's record of completed recall work, confirm with the manufacturer by VIN, and book any open recall straight after you buy, as the repair is normally free. See buying a used car with an outstanding recall for a printable checklist.
Frequently Asked Questions
Which Car Has The Fewest Recalls?
It depends on how you count. See the per-100,000 rankings on which car makes have the most recalls.
Is A Model With Many Software Recalls Worse?
Not necessarily. Software fixes are often quick dealer or over-the-air updates. See software recalls.
Can I Compare Vans?
Yes. Light vans are included, but compare vans with vans. See van recalls.
Data: DVSA vehicle recall data (Open Government Licence v3.0), updated 15 October 2026. Licensed vehicle figures: DfT vehicle licensing statistics (2026 Q2).
Sources
- Models with fewer than 5 refs show lists, not rates.
- Fleet survival (DfT VEH0124, end-2025): Still licensed: Citroën C3 (first used 2008 to 17) ≈ 69,465; Citroën DS3 (2009 to 16) ≈ 64,469; DS 3 non-Crossback (2015 to 19) ≈ 29,373; C4 (2010 to 18) ≈ 22,819; DS4 ≈ 9,229; DS5 ≈ 5,026. Vauxhall Zafira (2005 to 14) ≈ 92,375; Ford Kuga PHEV ≈ 48,302; Mazda MX-30 ≈ 6,528; Dacia Spring ≈ 6,018. 43 % of licensed cars were first used in 2015 or earlier; 17 % in 2010 or earlier.
- Code timelines: Producers normally respond to DVSA within 28 days (5.5); DVSA reviews are typically monthly (5.4). DVSA must be notified "as soon as the safety defect has been confirmed" (6.1). 3 notifications per motorist is seen as reasonable (8.3). Response rates are reported every three months (11.1).
- Official DVSA closed-recall response rates (SAF/04): cars & LGVs 89.9 % (2013-14), 91.4 % (2014-15), 89.0 % (2015-16), 98.0 % (2016-17).
- carVertical study (history reports, Jan 2023, Sep 2024, 26 countries): The UK had the lowest recall rate (3.1 % of cars recalled at least once). 72 % of recalled UK cars were still unrepaired. Commercial study, methodology not public. am-online.com, fleetnews.co.uk Checked 15 Oct 2026.
- Shared-platform recalls: DVSA refs listing several makes/models (e.g. Stellantis 1.2 PureTech across Peugeot/Citroën/Vauxhall/DS; JLR codes shared by Jaguar and Land Rover) are flagged in compare views.
- DVSA via Trading Standards (19 Feb 2018): "By law, car dealers, including online sellers, must fix safety recalls free of charge before selling cars to consumers. Private sellers aren't bound by the same rules." About 1 in 13 vehicles has an outstanding safety recall. In 2017, 87,700 of 1.14 million MOT-history recall enquiries found an outstanding recall. The recall check covered 92 % of cars and small vans. tradingstandards.uk