How long is a PhD programme and how long do students take?

Direct answer: A UK university publishes three numbers for one degree: a minimum period before you may submit, an expected period, and a maximum period after which registration lapses. At the University of Surrey (2026) a full-time PhD runs 30, 36 and 48 months respectively.

Asking how long is a PhD programme at one university shows why a single headline number misleads. The University of Surrey regulations for 2026-27 (PDF) list a full-time Doctor of Philosophy with a minimum registration of 30 months, an expected period of "36 months or within the duration of research funding if longer", and a maximum of 48 months (University of Surrey, 2026, para. 22). The part-time route carries 48, 72 and 96 months.

Source: University of Surrey, regulations for research degrees 2026-27, read 2026-10-31.

The expected period is the number a prospectus repeats, the minimum is the earliest your examiners can see a thesis, and the maximum is the one that ends your enrolment. Planning against the first while the university enforces the third is the commonest way to be surprised. The same regulations let the Admission, Progression and Examination Sub-committee reduce the minimum when your supervisors and the Pro-Vice-Chancellor support the case (University of Surrey, 2026, para. 23), so a two-and-a-half-year route is a permission, not a right.

Direct answer: Longer. Van de Schoot et al. (2013) surveyed Dutch candidates and found an average actual duration near 60 months, about 10 months past the contract. A UKRI-commissioned review (SQW, 2025) estimates that only one student in four or five submits inside the funded period.

Van de Schoot et al. (2013) is the clearest measurement of the gap because the authors asked each candidate for both the contract length and the real finishing time. Their first study had 565 respondents (a 50.7% response rate), of whom 301 candidates were analysed. Women averaged 59.8 months, with a mean delay of 9.52 months, and men averaged 59.67 months, with a mean delay of 10.11 months. Their second study asked 2,870 candidates in the Netherlands to complete a questionnaire; 1,504 did, and of the 232 candidates in a final year whose stage was known, 88 (38%) expected problems finishing on time. The paper notes that completion in the Netherlands is "around 75 per cent" in general.

For the UK, no national figure is published in the regulations themselves, so the best public signal is indirect. SQW, writing for UKRI, states that "submission within the funded period is relatively infrequent, estimated at around one-fifth to one-quarter of students" (SQW, 2025, para. 7.9). That sentence is an estimate in a report about stipends, not a cohort statistic, which is why it is worth reading as a direction rather than a measurement. Both sources answer how long is a PhD programme for the person paying and for the person writing, and they point the same way: the funded clock tends to stop before the thesis does.

Direct answer: Roughly half to two thirds finish within ten years in the large US project. Sowell et al. (2008) reported that 57% of entrants completed within that span, from 64% in engineering down to 49% in the humanities, and the social sciences kept rising after year ten.

The Council of Graduate Schools ran the largest public completion study. Its team wrote that "57% of the doctoral candidates in the sample completed their degree programs within a ten-year time span" (Sowell et al., 2008, p. 4). The demographic follow-up covered entrants from 1992-93 to 1994-95 at 24 institutions and counted 9,683 students in the gender database. By year ten 58% of men and 55% of women had finished, a gap of about three percentage points that the authors tested as statistically significant. By broad field the ten-year rate ran from 64% in engineering to 49% in the humanities. Science, engineering and mathematics fields reached 59% against 53% for social sciences and humanities combined, and the second group was still climbing at year ten, so its final rate is higher than the ten-year figure shows.

Two cautions follow. These are American programmes where coursework sits ahead of the dissertation, so the years do not transfer to the UK. And a 57% ten-year rate describes a cohort, not your own odds, which depend on funding, supervision and design more than on the field label. The dataset is also older than the other studies here, so the rates are useful for their shape rather than their exact level today.

Direct answer: Part-time enrolment lengthens it by about 0.7 years in South African national records, and publishing four or more articles was linked to a shorter time at one Australian university. Both are associations in observational data, so neither proves what would happen to you.

Van Lill (2024) analysed 3,551 doctoral records from 26 South African public universities across five disciplines between 2000 and 2021 and reported a mean time-to-degree of 4.25 years. The abstract states that "mode of enrolment is the strongest predictor of completion time" (Van Lill, 2024, p. 168), with part-time students taking on average 0.7 years longer than full-time students. That is elapsed time, not a judgement on effort.

Churchill et al. (2021, doi:10.28945/4875) examined 1,143 graduates of one Australian university, with time-to-degree data on 664. Median time fell from 4.13 full-time-equivalent years for graduates with no publications to 4.00 for one to three and 3.76 for four or more; the share finishing inside 3.99 years rose from 42% to 46% to 66%. The authors conclude that "Graduates with the highest research productivity (four or more articles) also exhibit the shortest time-to-degree" (Churchill et al., 2021, p. 700). Candidates in a health-discipline cohort programme recorded a median of 3.35 years against 3.97 outside it. Because productive students may also be the faster ones, treat this as a pattern worth knowing, not a recipe.

Direct answer: You apply for an extension to registration, and the university decides. At Surrey the extension is capped at 12 months in aggregate, and a student who reaches the maximum period without one has registration terminated as lapsed. Funding is a separate clock and does not extend automatically.

Surrey's rules are specific. A student "may apply for an extension to the maximum period of registration", the maximum extension is 12 months aggregate, and without one "their registration will be terminated on the grounds that it has lapsed" (University of Surrey, 2026, para. 26). Applied to the full-time PhD above, the last permitted month is month 60, and only if the committee agrees. The funder is a different decision-maker: a studentship that ends at 36 or 42 months pays nothing in month 45, while the university may still hold your enrolment open.

The gap in the data becomes personal at this point. If a typical candidate lands between the expected period and the maximum, and funding follows the expected period, the unfunded months sit at the end, when living costs continue but the stipend has stopped. Our guide to who decides PhD studentship funding explains which body holds each clock in the UK, and the backwards timeline for a PhD application shows how early those questions need an answer.

Direct answer: Convert everything to months, line the figures up, and plan for the median rather than the term. A full-time PhD with a 36-month expected period and a 48-month maximum sits close to the 49.6-month median for graduates with no publications in Churchill et al. (2021).

Put the figures in one table before you pick a programme.

The conversions multiply years by 12. The institutions differ and FTE years are not calendar months, so read the table as a spread of plausible outcomes, not as one prediction.

Three questions follow. Does the funding last as long as the middle of that range, or only the expected period? Does your design need ethics approval, fieldwork or restricted archives that add waiting time? And is part-time study a real choice, given Van Lill's 0.7-year difference? Whether you are finishing a degree full-time or working alongside it, our guide to part-time PhD study while working sets out how the ceilings change.

If you are choosing programmes and want a second opinion on which clock applies to you, PhD Application Strategy pairs you with an expert who reviews your plan while you write your own proposal. The expert can help you read a programme's minimum, expected and maximum periods against a funder's term and against the design of the research you want to do, so that the question of how long is a PhD programme gets an answer built for your case. The service sets no grade targets, makes no promise about admission or completion times, and if the match is not right you can ask for a different expert. Your proposal, your data and your thesis stay your own work throughout, and the expert's role is to ask the questions a supervisor would ask before you commit.

• Who decides PhD studentship funding?: which body pays for which months of the programme
• Part-time PhD while working: how part-time ceilings and funding change the plan
• PhD application timeline, worked backwards: when to settle funding and supervision before applying

Churchill, M. P., Lindsay, D., Mendez, D. H., Crowe, M., Emtage, N., & Jones, R. (2021). Does publishing during the doctorate influence completion time? A quantitative study of doctoral candidates in Australia. International Journal of Doctoral Studies, 16, 689-713. https://doi.org/10.28945/4875

Sowell, R., Zhang, T., Bell, N., & Redd, K. (2008). Ph.D. completion and attrition: Analysis of baseline demographic data from the Ph.D. Completion Project. Council of Graduate Schools. https://cgsnet.org/wp-content/uploads/2022/01/phd_completion_and_attrition_analysis_of_baseline_demographic_data-2.pdf

SQW. (2025). Understanding the effect of UKRI stipend levels on UKRI studentships. UK Research and Innovation. https://www.ukri.org/wp-content/uploads/2025/01/UKRI_290125-SQW-Understanding-the-effect-of-UKRI-stipend-levels-on-UKRI-studentships.pdf

University of Surrey. (2026). Regulations for research degrees 2026-27. https://www.surrey.ac.uk/sites/default/files/2026-08/A2-regulations-for-research-degrees-2026-27.pdf

Van de Schoot, R., Yerkes, M. A., Mouw, J. M., & Sonneveld, H. (2013). What took them so long? Explaining PhD delays among doctoral candidates. PLoS ONE, 8(7), e68839. https://doi.org/10.1371/journal.pone.0068839

Van Lill, M. H. (2024). A study of doctoral time-to-degree in selected disciplines at South African public universities. South African Journal of Higher Education, 38(1), 168-188. https://doi.org/10.20853/38-1-6268

This article is part of the MAAS Journal series for students. MAAS is an academic support partner: experts guide students through each stage of their own work, and we do not complete assessed work on a student's behalf.