Are UK driving tests easier in some months than others?

Section 1, The big finding

Across 2022-23 to 2024-25, the UK national pass rate moved between 48.0% in May and 49.3% in December. A spread of 1.3 percentage points between the kindest and the cruellest month of the calendar year.

I went into this expecting June, July and August to top the chart. Bright mornings, dry tarmac, students between exam terms with the time to practise. The data disagrees. The three lowest pass-rate months in the calendar year are April, May and June. The three highest are July, August and December. The summer-is-easier intuition has it roughly inverted.

None of this is a giant effect. The spread is barely over a percentage point. To put it in perspective, the gap between UK centres is over 35 percentage points, and the gender gap is around 4. Seasonality is small compared to where you book. It is real, though, and bigger than most learners would guess in either direction.

Section 2, How seasonality differs by country

The national average smooths over real regional differences. The three GB countries follow recognisably different patterns once you split them out. Northern Ireland is not included because DVSA's quarterly DRT122A release does not cover NI test centres, and the equivalent DVA release is published in a different format we do not currently ingest.

England follows the national pattern because England is most of the national sample, around 87% of the tests in this window. April-June dip; July, August and December are the high months.

Scotland runs lower overall (average around 47.2%) and its calendar-month shape is messier. July is still the high month but the seasonal swing is dominated by very small island and Highland centres where a handful of extra passes in any one month moves the figure by several percentage points. Read the Scotland chart with that caveat in mind.

Wales is the most seasonal of the three, with a spread of 2.3pp between its best and worst month. The Welsh pattern peaks in October and dips around the spring, which is loosely consistent with a tourist-season story (the Welsh test population skews more toward coastal centres that quieten down outside school holidays), but the Welsh sample is also smaller so confidence intervals are wider. We would not bet the farm on the Welsh shape persisting into the next DVSA release.

Northern Ireland is missing from this analysis. The Driver and Vehicle Agency in Northern Ireland publishes its own statistics, separately from DVSA, and we do not yet ingest that data. NI candidates should treat the national chart above as suggestive rather than directly applicable.

Section 3, Why this happens (cautious analysis)

We can describe the pattern with confidence. Why it happens is harder. Here are the plausible drivers, each labelled honestly: what the data supports, what it does not.

Hypothesis 1, the booking surge in spring (supported)

DVSA does not publish per-month booking volume per centre, only tests taken. But the test-volume series shows a clear rise from January through May, peaking in late spring as the warm-weather effect kicks in. More candidates takes the average down because the marginal new candidate is, on average, less prepared than the steady-state caseload. The April-June dip is consistent with this. We cannot prove it without seeing booking data, but the test-volume series in our dataset rhymes with it.

Hypothesis 2, school holidays (mixed)

July and August coincide with the UK summer school holidays. A learner who has finished GCSEs or A-levels has six weeks to cram lessons. The 17-18 cohort is a large slice of new candidates and they pass slightly better than the all-ages average. That would push the summer months up. The data shows this for July and August, less clearly for September. It is consistent with the school-holiday hypothesis but does not prove it, because we cannot split pass rates by age in DRT122A.

Hypothesis 3, weather (weak)

Folk explanation: people fail in winter because the roads are slippery and the examiner is grumpy. The data does not show this. December has one of the highest pass rates in the calendar year, not the lowest. Tests cancelled outright for ice or fog do not appear in the dataset at all, so the December figure is a survivorship-biased estimate: only tests that did go ahead are counted, and those went ahead because the weather was tolerable on that specific morning. We cannot tell from this data whether weather actually affects pass rate, only that the December candidates who actually take the test pass at a high rate.

Hypothesis 4, examiner staffing (speculative)

DVSA examiner staffing levels rise and fall with annual leave patterns. Summer holiday absences could shift the examiner mix at any one centre, with newly-deployed examiners on average slightly stricter. We have no way to confirm or deny this from the public dataset. It is the kind of thing an FOI request could clarify, and a useful direction for a follow-up post.

Section 4, What learners should do with this

Even though the seasonal effect is real, it is small. A learner who can pick their booking month should not over-weight the seasonal angle relative to the much larger effects below.

Things that matter more than which month you book

• Which centre you book. Same-day pass-rate spread between UK centres is over 35 percentage points (easiest centres vs hardest centres). Picking the right centre dwarfs picking the right month by at least 25x.
• How prepared you are. First-time pass rate correlates with hours of professional instruction. DVSA itself suggests around 45 hours of lessons plus 22 hours of private practice as a typical figure. See our lessons guide.
• Whether it is your first attempt. First-time and overall pass rates differ by around 5 percentage points nationally and up to 15 at individual centres. Some "easy" centres look easy because their retake pool inflates the headline figure. See first-time pass-rate explainer.

If everything else is equal

The rational play, holding centre and preparation constant, is to book in July, August or December rather than April, May or June. We are talking about an expected lift of roughly half a percentage point. For a learner whose true probability of passing sits at 60%, that nudges to around 60.5%. Detectable in the aggregate, swamped by individual variation in any one candidate.

A common misconception worth flagging: many learners think they should book the first available test regardless of month, on the grounds that the wait list is so long that month-shopping is impossible. That is empirically not always true. The DVSA booking system does offer non-trivial choice of week and month for many centres, and a learner willing to wait an extra fortnight for a better-month slot loses less than a half-percentage point of expected probability. That is a small price.

Section 5, Methodology and limitations

Data source. DVSA DRT122A quarterly release, financial years 2022-23 to 2024-25. Each row gives a centre-month-gender combination with test count and pass count. We aggregate to the all-gender total per centre per month, then sum across centres into UK and per-country tables. All figures licensed under the Open Government Licence v3.0.

Volume-weighted aggregation. Pass rate for a month is sum of passes across all centres divided by sum of tests, not the average of per-centre pass rates. This means a busy centre like Mitcham or Wolverhampton contributes proportionally more to the figure than Lerwick. This is the same convention DVSA uses for its own quarterly press release headlines, and the same convention used throughout PassRates.uk. Our full method is documented on the methodology page.

Sample threshold. Centre-months with fewer than 5 tests in a cell are redacted by DVSA in the source spreadsheet to ".." (disclosure control). We skip these redacted rows in the aggregation because they carry no usable volume to weight by. Across the three-year window, redacted rows account for less than 0.1% of test volume, so the effect on the headline figures is negligible. The redacted-row count climbs at small island centres, which is one reason the Scotland chart should be read with wider confidence bands than the England chart.

Centres counted. 294 car test centres contributed to this analysis. Centres in the KNOWN_CLOSED set, motorcycle and lorry centres, and centres with no recent data are excluded, matching the same filter the rest of PassRates.uk uses. Riding test "(R)" centres are excluded too.

What we did not do. We did not adjust for COVID disruption. The three-year window from 2022-23 onwards sits after the post-lockdown booking surge, so the lingering COVID effect is small but non-zero (a longer wait list pushes the candidate pool toward more prepared candidates, all else equal). We did not split by age, by first attempt vs retake, or by examiner. We did not include Northern Ireland. Each of those is a future research direction. If you spot a gap or a mistake, email hello@passrates.uk and we will correct it.

Cite this page: passrates.uk research/seasonality v1.0 (2026). Underlying data: DVSA DRT122A (OGL v3.0), 2022-23 to 2024-25.

Quarterly summary, at a glance

Calendar quarter beats calendar month if you want a decision-relevant number: it pools enough volume to swamp noise and matches how learners actually plan ("I want to pass before Christmas", "Before I head to uni").

Related on PassRates.uk

• How we calculate pass rates
• Easiest UK test centres
• Hardest UK test centres
• Understanding pass rate statistics
• Should I travel for an easier test centre?