Practical driving test pass rate — and what people actually fail on

The car practical driving test pass rate in Great Britain for April 2025 to March 2026 was 50.0% — 1,000,043 passes from 1,998,621 tests. That is the highest figure in the series DVSA publishes, which runs back to April 2007, and the first year it has reached 50%. The gap between the best and worst test centre is more than 30 percentage points, but almost none of that gap is about the examiner: what separates a pass from a fail is a short, published, repetitive list of faults.

What people actually fail on

DVSA records every fault an examiner marks and publishes the ten most frequently recorded ones each year in table DRT121F. The list barely moves. Junction observation has been number one in every year the table has been published, back to 2007/08; mirrors before changing direction has been number two in every year since 2008/09; and eight of this year's ten also appear in the 2015/16 list.

Six of the ten are observation or response faults, not car-control faults. Learners tend to revise the manoeuvres and worry about stalling; the marking sheet is dominated by whether you looked, and whether you acted on what you saw. Year on year the list is close to frozen: nine of the ten entries are identical to 2024/25, reordered, and the only substantive change is at number 10, where use of speed has been replaced by maintaining progress — speed. That one covers driving too slowly and waiting at a junction when the gap was there.

Driving, serious and dangerous: what moves a fault up a band

Three bands, and only one of them is survivable. The pass threshold is fixed and public:

You’ll pass your driving test if you make: no more than 15 driving faults (sometimes called ‘minors’); no serious or dangerous faults (sometimes called ‘majors’).

The band is not a property of the manoeuvre — it is a property of the risk. The same fault at the same junction can be marked in any of the three bands, and DVSA publishes the worked example: misjudging the distance of an approaching vehicle after taking observation is a driving fault; not taking effective observation at all and emerging into the path of other vehicles is serious; the same lapse where the examiner has to act to avoid a collision is dangerous.

Repeating one driving fault is how most “0 majors” fails happen

A single driving fault never fails anyone. Doing the same one all the way round can:

In isolation a driving fault does not result in failure to pass the test. However a candidate who habitually commits a driving fault in one aspect of driving throughout the test, cannot be regarded as competent to pass the test. The repeated fault can then be assessed as potentially dangerous.

Habitually failing to check the mirrors when it matters is the example DVSA gives. That is why a debrief can end with a serious fault for something that felt small each time it happened — the escalation is in the pattern, not in the individual event.

A breach of the law is automatically serious

Not responding to a red light, driving well above the speed limit and not responding to a stop sign are all listed by DVSA as breaches of a legal requirement, and a breach of that nature is a serious fault. “Nothing happened” is not a defence: the test measures the risk you created, not the outcome you got away with.

The two questions the examiner is actually asking

DVSA publishes the reasoning examiners are trained to apply, which is more useful to a learner than any list of tips. The first is the first rule of assessment:

If the drive was perfect before and perfect after this fault occurred, would this fault and this fault alone, cause the candidate to fail their test?

Each fault is weighed on its own, not against how the rest of the drive went. The second is the 16th driving fault rule: before recording anything, the examiner asks whether, if this were the 16th fault, it could be justified as the one that failed you. If not, it is marked “not worthy” and never appears. Dry steering, sequential gear changes and not pressing the handbrake button are the examples DVSA gives of faults not worth recording at all.

Why the examiner keeps driving after you have already failed

Because the length of the test is set by regulation, not by how it is going. For a category B car test the on-road driving time is not less than 30 minutes, measured from the moment the car starts moving to the moment the engine is switched off, and DVSA puts the total test at 38 to 40 minutes. Minimum driving times are strictly adhered to unless the test is terminated.

Examiners are also instructed to avoid telling you that you have failed before the end. If you ask outright, they will tell you — but the guidance is explicit about what usually happens next:

Divulging a failure can demoralise the candidate and/or cause further poor driving or an insistence they don’t want to go any further.

Asking to stop ends the test there. It is recorded as a terminated test and the car goes no further than a safe place to pull up — which can leave you a long way from the test centre. The drive is worth finishing anyway: whatever the result, the examiner tells you what faults you made, and that debrief is the most specific practice list you will ever be handed. The exception runs the other way — if the driving becomes dangerous enough to threaten public safety the examiner stops the test, issues a statement of failure and says why.

The numbers behind the 50%

Three of those deserve a second look. First attempts pass slightly more often than the average (50.5% against 50.0%), which is the opposite of the folk wisdom that everyone fails the first one — the average is dragged down by the pool of repeat retakes, not by first-timers. A clean sheet is genuinely rare: about one pass in twenty is a zero-fault pass, so 1 to 15 driving faults is the normal shape of a pass, not a near miss. And automatic tests pass at 44.6% against the 50.0% overall — a gap that has held every year since the series began, on a share of tests that has grown to 29% of all car tests in 2025/26. DVSA publishes the two rates and not the reason for the gap, and an automatic-only cohort is not a like-for-like comparison, so read it as a fact about the population rather than about the gearbox.

The gender split also runs the opposite way to the theory test, where female candidates passed at 47.2% against 43.0% for male candidates in the same year. See the theory test pass rate for that side of the comparison.

Ten-year trend

• 2016-17 47.1%
• 2017-18 46.3%
• 2018-19 45.8%
• 2019-20 45.9%
• 2020-21 49.8%
• 2021-22 48.9%
• 2022-23 48.4%
• 2023-24 47.9%
• 2024-25 48.7%
• 2025-26 50.0%

The practical pass rate has been remarkably stable — a band of 44% to 50% across nineteen published years. That is a very different shape from the theory test, which fell from 65% in 2007/08 to the mid-40s today.

Pass rate at every test centre

DVSA table DRT122A reports pass rates at 317 car driving test centres for April 2025 to March 2026. Search by name, sort by any column. Among centres conducting at least 1,000 tests, the highest are Melton Mowbray at 69.7% (1,984 tests), Barrow In Furness 69.6% (1,699) and Dorchester 69.0% (4,353); the lowest are Wolverhampton at 34.3% (11,561), Wednesbury 35.0% (7,508) and Barking (Tanner Street) 36.3% (5,543). Sort by rate and the top row is actually Isles of Scilly at 75.0% — six passes from eight tests all year, which is the whole problem with reading the smallest centres at face value.

All 317 centres — search and sort

Search test centres by name

317 centres

Centre ↕	Tests ↕	Passes ↕	Pass rate ▼	YoY (pp) ↕
Isles of Scilly	8	6	75.0%	-12.5
Isle of Mull	42	31	73.8%	+2.4
Gairloch	49	36	73.5%	+4.3
Pitlochry	146	107	73.3%	—
Benbecula Island	107	78	72.9%	-5.9
Girvan	250	180	72.0%	+3.3
Campbeltown	109	78	71.6%	+9.0
Dunoon	193	138	71.5%	+8.7
Stranraer	215	153	71.2%	-2.2
Golspie	93	66	71.0%	+3.0
Duns	171	120	70.2%	-4.0
Montrose	621	435	70.0%	+3.3
Melton Mowbray	1,984	1,382	69.7%	+5.8
Barrow In Furness	1,699	1,183	69.6%	+7.7
Inveraray	124	86	69.4%	-15.7
Dorchester	4,353	3,002	69.0%	+2.3
Aberfeldy	149	102	68.5%	—
Pwllheli	737	502	68.1%	+2.1
Lerwick	379	257	67.8%	0.0
Arbroath	605	408	67.4%	-11.5
Cumnock	760	512	67.4%	+6.3
Brodick (Isle of Arran)	55	37	67.3%	—
Stornoway	344	229	66.6%	+10.5
Oban	322	213	66.1%	+4.8
Newton Stewart	225	148	65.8%	+2.6
Fort William	374	245	65.5%	+3.5
Kendal (Oxenholme Road)	2,231	1,453	65.1%	+0.3
Tunbridge Wells	5,726	3,695	64.5%	+5.6
Hawick	415	267	64.3%	+3.2
Malton	876	563	64.3%	-2.4
Abergavenny	5,924	3,805	64.2%	+3.1
Mallaig	53	34	64.2%	+1.2
Fraserburgh	531	340	64.0%	+3.3
Huntly	245	156	63.7%	+1.2
Yeovil	7,493	4,752	63.4%	+3.7
Lee On The Solent	8,502	5,374	63.2%	+3.6
Alness	783	494	63.1%	+0.1
Thurso	162	102	63.0%	+4.9
Ipswich	11,466	7,143	62.3%	-0.8
Bangor	4,032	2,509	62.2%	-1.9
Inverurie	749	464	61.9%	+7.6
Workington	2,998	1,857	61.9%	+13.6
Norwich (Peachman Way)	12,541	7,740	61.7%	+4.7
Bala	607	374	61.6%	+0.5
Chichester	6,672	4,109	61.6%	-2.6
Ullapool	49	30	61.2%	-13.4
Weston-super-Mare	4,443	2,711	61.0%	+4.5
Northallerton	3,244	1,958	60.4%	+1.8
Forfar	667	402	60.3%	-8.2
Maidstone	8,053	4,850	60.2%	+2.2
Wellingborough	1,450	870	60.0%	+4.9
Walton LGV	1,894	1,132	59.8%	+2.9
Worthing	8,495	5,076	59.8%	+1.0
Farnborough	10,482	6,258	59.7%	+1.4
Taunton	8,259	4,915	59.5%	+3.2
Wick	239	142	59.4%	+5.6
Heysham	3,692	2,187	59.2%	+4.2
Kelso	434	257	59.2%	+2.2
Kingussie	93	55	59.1%	-5.0
Sevenoaks	6,740	3,978	59.0%	+5.4
Hereford	4,110	2,416	58.8%	-0.6
Peebles	685	403	58.8%	-8.2
Brecon	1,490	875	58.7%	-0.2
Buxton	4,904	2,878	58.7%	+1.7
Carlisle	3,518	2,065	58.7%	+3.8
Ludlow	5,051	2,966	58.7%	-0.9
Sidcup (London)	14,318	8,405	58.7%	-0.3
Barnsley	7,737	4,530	58.5%	+6.3
Knaresborough	8,301	4,860	58.5%	0.0
Lanark	1,914	1,112	58.1%	+3.1
Newtown	2,256	1,309	58.0%	-5.7
Grantham (Somerby)	7,318	4,237	57.9%	-0.2
East Kilbride	1,976	1,140	57.7%	+11.3
Northwich	3,529	2,032	57.6%	+2.9
Ayr	2,189	1,258	57.5%	+0.7
Banff	280	161	57.5%	+6.9
Haddington	2,195	1,263	57.5%	-4.7
Letchworth	5,243	3,015	57.5%	-0.8
Kyle of Lochalsh	136	78	57.4%	-1.1
Ashfield	8,144	4,666	57.3%	+6.1
Dundee	3,742	2,144	57.3%	-3.0
Skipton	3,197	1,831	57.3%	+2.9
Guildford	4,542	2,592	57.1%	+5.3
Barry	2,244	1,280	57.0%	-4.8
Isle of Skye (Portree)	128	73	57.0%	-12.9
Grantown-On-Spey	137	78	56.9%	-5.3
Rhyl	4,011	2,284	56.9%	+1.0
Basingstoke	5,009	2,844	56.8%	-0.1
Whitby	557	316	56.7%	-8.4
Birmingham (Shirley)	9,385	5,309	56.6%	-1.5
Greenham	3,931	2,220	56.5%	+4.4
Southport (Liverpool)	7,949	4,493	56.5%	-2.7
Chippenham	9,629	5,435	56.4%	+5.8
Gillingham	14,886	8,389	56.4%	-0.8
Lochgilphead	140	79	56.4%	-6.1
Preston	9,194	5,174	56.3%	+2.6
Chesterfield	6,605	3,712	56.2%	+1.7
Bradford (Heaton)	8,354	4,690	56.1%	+6.6
Bredbury (Manchester)	10,636	5,953	56.0%	+1.8
Herne Bay	10,674	5,970	55.9%	+0.5
Orkney	320	179	55.9%	-10.1
Skegness	2,078	1,162	55.9%	-1.7
Irvine	4,941	2,752	55.7%	+6.5
Livingston	2,997	1,669	55.7%	+5.0
Alnwick	1,282	712	55.5%	+4.5
Bridlington	3,027	1,681	55.5%	-0.9
Bristol (Avonmouth)	9,400	5,214	55.5%	+1.8
Warrington	7,550	4,193	55.5%	+0.2
Chelmsford	9,107	5,044	55.4%	—
Eastbourne	6,031	3,342	55.4%	+3.3
Reading	9,237	5,118	55.4%	+5.6
Stevenage	4,622	2,557	55.3%	+10.4
Trowbridge	3,074	1,698	55.2%	+4.1
Worksop	5,975	3,296	55.2%	+1.9
Aberystwyth (Park Avenue)	1,529	843	55.1%	+1.9
Burton on Trent	7,273	3,981	54.7%	+0.6
York	9,416	5,125	54.4%	+3.1
Wallasey	968	526	54.3%	+4.7
Carmarthen	2,746	1,488	54.2%	-0.2
Beverley LGV	159	86	54.1%	-1.0
Newport (Isle of Wight)	3,616	1,958	54.1%	+1.7
Llantrisant	9,909	5,352	54.0%	+3.4
Berwick-On-Tweed	666	359	53.9%	-0.4
Dumfries	2,267	1,223	53.9%	+6.9
Enfield (Innova Business Park)	8,412	4,536	53.9%	-0.2
Louth	740	399	53.9%	-0.5
Redhill Aerodrome	5,408	2,916	53.9%	+4.5
Cambridge (Brookmount Court)	13,689	7,366	53.8%	+3.3
Newton Abbot	6,591	3,547	53.8%	+1.9
Bedford	8,854	4,756	53.7%	+3.8
Tolworth (London)	10,164	5,447	53.6%	+0.3
Birmingham (Cocks Moors)	486	260	53.5%	—
Castle Douglas	549	293	53.4%	+3.2
Poole	10,375	5,538	53.4%	+5.0
Bridgend	5,990	3,193	53.3%	+0.2
Doncaster	7,334	3,903	53.2%	+11.2
Durham	9,451	5,027	53.2%	+3.6
Nuneaton	6,764	3,600	53.2%	+0.7
Peterhead	2,628	1,399	53.2%	-4.6
Rotherham	9,134	4,854	53.1%	+1.8
Hexham	1,388	735	53.0%	-3.7
Lichfield	5,944	3,153	53.0%	+4.4
Bromley (London)	13,228	6,997	52.9%	+1.2
Chertsey (London)	9,082	4,804	52.9%	+2.6
Chorley	5,415	2,860	52.8%	+4.5
Dumbarton	1,659	876	52.8%	-0.7
Peterborough	10,551	5,569	52.8%	+2.6
Portsmouth	13,518	7,141	52.8%	+1.0
Steeton	4,784	2,524	52.8%	+2.3
Yeading (London)	6,735	3,555	52.8%	+5.9
Upton	10,772	5,667	52.6%	+2.6
Aberdeen South (Cove)	4,858	2,546	52.4%	+1.0
Hastings (Ore)	3,892	2,039	52.4%	+3.2
Oswestry	1,851	969	52.4%	-3.2
Uxbridge (London)	6,051	3,163	52.3%	+2.7
Winchester	4,602	2,404	52.2%	+4.3
Lincoln	6,192	3,225	52.1%	+2.3
Northampton	13,539	7,050	52.1%	+1.7
Rothesay	125	65	52.0%	—
Bristol (Kingswood)	14,863	7,719	51.9%	-0.1
Kings Lynn	6,694	3,477	51.9%	+5.9
Macclesfield	4,509	2,341	51.9%	+2.0
Worcester	11,590	6,019	51.9%	+1.3
Wrexham	4,819	2,501	51.9%	-4.9
Blackpool	8,948	4,633	51.8%	+2.2
Bolton (Manchester)	11,570	5,996	51.8%	-4.9
Derby (Alvaston)	8,086	4,186	51.8%	+0.5
Leeds (Colton Mill)	6,432	3,332	51.8%	—
Leeds (Fearnville)	4,676	2,422	51.8%	—
Nelson	5,955	3,081	51.7%	+3.8
Erith (London)	5,809	3,000	51.6%	+2.3
West Wickham (London)	12,249	6,320	51.6%	+0.7
Bletchley	9,415	4,845	51.5%	+5.8
High Wycombe	3,534	1,821	51.5%	+0.5
Barnstaple	5,562	2,859	51.4%	-3.3
Hendon (London)	11,086	5,695	51.4%	+1.9
Mill Hill (London)	9,797	5,038	51.4%	+0.3
Redditch	4,483	2,304	51.4%	+4.7
Rugby	6,664	3,425	51.4%	+5.4
Sale (Manchester)	10,921	5,613	51.4%	+2.3
Bishops Stortford	6,409	3,290	51.3%	+0.9
Leicester (Wigston)	18,868	9,666	51.2%	+7.3
Merthyr Tydfil	4,446	2,275	51.2%	-2.3
Buckie	268	137	51.1%	-1.5
Sunderland	10,436	5,338	51.1%	+4.0
Swindon	12,857	6,567	51.1%	+1.8
Hartlepool	5,847	2,984	51.0%	+1.3
Llanelli	4,380	2,233	51.0%	0.0
Monmouth	1,283	654	51.0%	-10.2
Morden (London)	15,770	8,036	51.0%	+2.2
Shrewsbury	5,064	2,582	51.0%	+1.8
Blyth	6,973	3,551	50.9%	+3.7
Kirkcaldy	4,313	2,194	50.9%	+3.4
Borehamwood (London)	7,195	3,652	50.8%	+4.0
Slough (London)	10,509	5,343	50.8%	+5.2
Boston	4,784	2,427	50.7%	0.0
Grangemouth	3,761	1,906	50.7%	+4.3
St Albans	7,336	3,721	50.7%	+2.3
Colchester	7,756	3,925	50.6%	+2.6
Oxford (Cowley)	5,419	2,741	50.6%	+2.4
Isleworth (Fleming Way)	12,687	6,388	50.4%	-1.4
Tilbury	5,083	2,560	50.4%	+1.3
Southampton (Maybush)	10,480	5,273	50.3%	+3.2
Pembroke Dock	1,769	887	50.1%	-7.8
Barnet (London)	12,397	6,182	49.9%	0.0
Callander	335	167	49.9%	+2.5
Leighton Buzzard (Stanbridge Road)	7,667	3,825	49.9%	-1.1
Nottingham (Chilwell)	9,415	4,697	49.9%	+6.0
Bishopbriggs	6,186	3,080	49.8%	-0.7
Horsforth	12,466	6,193	49.7%	-1.6
Pinner (London)	14,367	7,143	49.7%	-0.6
Edinburgh (Musselburgh)	6,850	3,387	49.4%	+2.4
Aylesbury	5,255	2,589	49.3%	+0.6
Huddersfield	10,563	5,198	49.2%	+0.8
Basildon	8,359	4,106	49.1%	+3.5
Kettering	11,876	5,829	49.1%	+3.1
Newport (Gwent)	8,227	4,021	48.9%	-2.1
Telford	4,963	2,429	48.9%	+5.0
Tottenham	9,844	4,812	48.9%	+1.0
Perth (Arran Road)	2,642	1,290	48.8%	+2.8
Sheffield (Middlewood Road)	10,442	5,085	48.7%	+1.2
Paisley	7,912	3,847	48.6%	+2.0
Birmingham (Kingstanding)	16,133	7,829	48.5%	+3.9
Brentwood (London)	5,791	2,809	48.5%	0.0
Salisbury	6,630	3,215	48.5%	-4.0
Warwick (Wedgenock House)	15,070	7,313	48.5%	+0.1
Gosforth	10,531	5,100	48.4%	+0.9
Rochdale (Manchester)	11,563	5,584	48.3%	+7.1
Aberdeen North	3,906	1,881	48.2%	+5.1
Clacton-on-Sea	4,840	2,334	48.2%	+0.8
Swansea	12,375	5,970	48.2%	-3.9
Hull	9,706	4,652	47.9%	+2.7
Bradford (Thornbury)	10,831	5,172	47.8%	+0.7
Gloucester	7,420	3,545	47.8%	-1.8
Middlesbrough	8,854	4,221	47.7%	-0.6
West Didsbury (Manchester)	17,806	8,496	47.7%	-3.2
Bodmin	6,392	3,034	47.5%	+1.4
Airdrie	4,599	2,180	47.4%	+6.1
Galashiels	2,241	1,063	47.4%	-1.7
Glasgow (Baillieston)	10,428	4,947	47.4%	+3.0
Chester	9,212	4,360	47.3%	+1.1
Darlington	12,183	5,740	47.1%	-2.1
Loughton (London)	7,376	3,465	47.0%	-1.0
Mitcham (London)	15,046	7,076	47.0%	-1.9
Southend-on-Sea	7,526	3,537	47.0%	-1.1
Pontefract	5,544	2,602	46.9%	-0.9
Birmingham (Kings Heath)	9,072	4,242	46.8%	-0.2
Chadderton	15,988	7,475	46.8%	+1.9
Exeter	8,679	4,061	46.8%	-1.1
Wood Green (London)	11,379	5,326	46.8%	0.0
Blackburn with Darwen	12,205	5,688	46.6%	-1.2
Cardiff (Llanishen)	16,401	7,634	46.5%	-4.8
Penzance	507	236	46.5%	-2.7
Hinckley	4,866	2,258	46.4%	-1.9
Nottingham (Colwick)	15,150	7,032	46.4%	+0.9
Bury (Manchester)	7,024	3,255	46.3%	+6.0
Hornchurch (London)	8,694	4,022	46.3%	-2.0
Launceston	3,991	1,849	46.3%	-2.0
Ashford (Kent)	5,475	2,530	46.2%	-5.8
Bury St Edmunds	8,702	4,019	46.2%	+0.5
Watford	12,939	5,983	46.2%	-0.4
Enfield (Brancroft Way)	362	167	46.1%	+3.7
Canterbury	4,670	2,138	45.8%	-4.5
Heckmondwike	5,471	2,505	45.8%	-0.9
Inverness (Longman Drive)	3,252	1,491	45.8%	-0.1
Stafford	5,365	2,459	45.8%	-0.3
Burgess Hill	11,237	5,111	45.5%	+2.3
Scunthorpe	6,103	2,773	45.4%	+1.0
Wakefield	11,653	5,291	45.4%	-1.3
Dudley	13,721	6,200	45.2%	-0.5
Scarborough	3,595	1,620	45.1%	-6.6
Halifax	5,845	2,629	45.0%	+0.3
Southall (London)	8,543	3,827	44.8%	-1.4
Carlisle LGV (Cars)	47	21	44.7%	-17.2
Cheetham Hill (Manchester)	10,207	4,565	44.7%	+1.0
Birmingham (Garretts Green)	22,480	9,980	44.4%	+2.4
Wanstead (London)	8,258	3,670	44.4%	+4.0
Atherton (Manchester)	13,035	5,779	44.3%	-0.1
Glasgow (Anniesland)	6,170	2,736	44.3%	+3.5
Sheffield (Handsworth)	9,788	4,316	44.1%	-1.2
Crewe	4,994	2,187	43.8%	-0.6
Goodmayes (London)	21,374	9,338	43.7%	0.0
Lowestoft(Mobbs Way)	7,603	3,292	43.3%	-2.8
Cheltenham	5,161	2,231	43.2%	-3.5
Stirling	2,747	1,182	43.0%	+4.3
Birmingham (South Yardley)	14,471	6,199	42.8%	+1.2
Grimsby Coldwater	5,361	2,292	42.8%	+0.1
Crawley	11,305	4,831	42.7%	+2.6
Widnes	6,846	2,922	42.7%	+2.6
Folkestone	7,818	3,313	42.4%	-3.0
Elgin	2,934	1,240	42.3%	-2.2
Loughborough	5,502	2,330	42.3%	-1.4
Hamilton	6,112	2,567	42.0%	+2.5
Greenock	3,547	1,479	41.7%	-0.7
Edinburgh (Currie)	6,202	2,577	41.6%	-1.6
Stoke-on-Trent (Newcastle-Under-Lyme)	6,646	2,768	41.6%	+0.9
Greenford (Horsenden Lane)	7,222	2,992	41.4%	+0.9
Banbury	9,069	3,741	41.3%	+1.4
Cardigan	1,192	489	41.0%	-8.8
St Helens (Liverpool)	14,174	5,803	40.9%	+0.7
Plymouth	11,167	4,546	40.7%	-0.1
Camborne	6,271	2,549	40.6%	-0.7
Stoke-On-Trent (Cobridge)	8,413	3,413	40.6%	+1.8
Belvedere (London)	4,245	1,720	40.5%	+2.2
Speke (Liverpool)	8,159	3,308	40.5%	+1.9
Leicester (Cannock Street)	14,198	5,708	40.2%	+2.5
Coventry	13,771	5,510	40.0%	-2.3
Chingford (London)	13,120	5,234	39.9%	+3.4
Glasgow (Shieldhall)	7,019	2,803	39.9%	+2.2
Luton	7,759	3,041	39.2%	-0.9
Norris Green (Liverpool)	13,472	5,099	37.8%	-0.9
Gateshead	9,512	3,568	37.5%	+0.1
Dunfermline (Vine)	5,529	2,012	36.4%	-6.9
Featherstone	13,193	4,807	36.4%	+2.3
Barking (Tanner Street)	5,543	2,011	36.3%	-1.6
Wednesbury	7,508	2,626	35.0%	-1.4
Wolverhampton	11,561	3,961	34.3%	+0.9

Biggest movers

204 of the 317 centres passed at least half their candidates in April 2025 to March 2026. A simple count of how many centres rose or fell is not worth much here: DVSA publishes rates to one decimal place, so a centre testing a few dozen candidates can cross the line on a single result. The swings below are therefore restricted to centres conducting 1,000 tests or more, where a move of this size reflects something real:

• Up: Workington +13.6 percentage points, East Kilbride +11.3, Doncaster +11.2, Stevenage +10.4
• Down: Monmouth −10.2 percentage points, Cardigan −8.8, Pembroke Dock −7.8, Dunfermline (Vine) −6.9

What the centre spread does and does not tell you

A 40-point spread looks like proof that some centres are stitching people up. Read the data carefully and it says something duller and more useful. DVSA attaches its own warning to the centre-level tables:

Low numbers within a particular data set must be treated with caution as they may not fully reflect the candidate characteristics being reported upon; any conclusions drawn will therefore not be statistically robust or reliable for any form of analysis or comparison e.g. comparing performance of a test centre where less than 100 tests were conducted against a test centre where over 1,000 were conducted.

Two things the table does establish, and two the table cannot:

• Volume noise is real. Isles of Scilly's 75.0% is six passes. Anything under a few hundred tests can move ten points on a handful of results, which is exactly what the note above warns about. Filter to 1,000 tests or more and the spread narrows from 40 points to 35.
• The urban–rural split is real. Sorted by rate with the small centres filtered out, the top is dominated by market towns and rural coastal centres and the bottom by dense urban centres in the West Midlands, outer London, Merseyside and central Scotland. That pattern holds across dozens of centres, not a handful.
• Why the split exists is not in the data. Route density and candidate mix are the usual explanations — an urban test route presents more junctions, crossings, lane changes and parked-car obstructions per mile, and every one of them is an opportunity to be marked — but DVSA publishes no measure of route difficulty, and reports only whether a test was a first attempt (table DRT122C) rather than the full attempt history by centre, so treat that as reasoning rather than measurement.
• It is not the examiners. Examiners are trained to a uniform standard set by the Chief Driving Examiner, and each fault is weighed on its own under the first rule of assessment above. Chasing a softer centre also means taking the test on roads you have never driven, which is the trade nobody making that plan prices in.

One thing the data genuinely cannot tell you: DVSA does not publish faults by test centre. The fault table is national and the centre table is pass rates only. Anyone showing you the top faults for your specific centre has estimated them.

How to be in the 50%

1. Practise the top four, not the manoeuvres. Junction observation and mirrors before changing direction have been the two most-recorded faults every year since 2008/09, with right turns and traffic lights close behind them this year. They are also the four you can rehearse on any drive.
2. Break the habits, not the one-offs. A driving fault you commit once costs you one of fifteen. The one you commit at every junction is the one that gets reassessed as serious. Ask your instructor which fault you repeat — that is the answer that changes results.
3. Treat legal requirements as untouchable. Red lights, stop signs and speed limits are not scored on outcome. Crossing one is a serious fault on its own.
4. Keep moving. Maintaining progress is back in the top 10 for 2025/26. Stopping and waiting when it is safe and reasonable to proceed is marked, and unlike a hesitant driver's instinct, it is not the safe option.
5. Finish the test. You cannot recover a serious fault, but you can walk away with a complete, examiner-written list of what to fix. A retake has to be booked for a date at least 10 working days away, so you have the time to use it.

All of that comes after the theory test — you can only book a practical once you have passed it. Our app has free DVSA revision questions, hazard perception clips, mock tests and the full Highway Code in one place. Work through the 14 DVSA syllabus topics, take a mock test and practise hazard perception separately. The theory pass mark and the theory pass rate are covered in full elsewhere on this site.

How DVSA publishes this data

Everything above comes from the Driving test and theory test data: cars collection on gov.uk, published as OpenDocument Spreadsheets:

• DRT121A — tests conducted, passed and pass rates by quarter and year
• DRT121E — tests passed with zero faults
• DRT121F — top 10 faults made
• DRT121G / DRT122F — waiting times and availability
• DRT122A — pass rates by gender and month at each test centre
• DRT122C — pass rates by first driving attempt at each test centre
• DRT122E — automatic-car pass rates

The centre-level table refreshes annually and was last updated in June 2026, covering April 2025 to March 2026. Centres where only one examiner conducted this category of test are left out of the centre-level table but still counted in the national totals, which is why the centre figures sum to slightly fewer tests than DRT121A reports. We update this page when each release lands.

Find your nearest centre with the official search at gov.uk/find-driving-test-centre, and book directly at gov.uk/book-driving-test — see our note on booking direct rather than through a middleman.

Sources

• GOV.UK — Driving test and theory test data: cars (DVSA tables DRT121A, DRT121E, DRT121F, DRT122A, DRT122C, DRT122E)
• GOV.UK — Driving test: driving test faults and result
• GOV.UK — Carrying out driving tests: examiner guidance (DT1) — section 11: guide to assessment and marking
• GOV.UK — Carrying out driving tests: examiner guidance (DT1) — section 1: car driving test
• GOV.UK — Carrying out driving tests: examiner guidance (DT1) — general technical matters
• GOV.UK — Driving test: booking your test

Test statistics and examiner guidance reproduced under the Open Government Licence v3.0.