Quick answer: In the UK, an undergraduate dissertation is usually 8,000–12,000 words (10,000 is the most common target) and a Masters dissertation is usually 12,000–20,000 words (15,000 is the most common target) — but your own handbook’s figure is the only one that counts, and it almost always allows a margin of plus or minus 10 per cent. A typical split for a 10,000-word project: introduction around 1,000, literature review 2,500–3,000, methodology 1,500–2,000, results or findings 1,500–2,000, discussion 1,500–2,000, conclusion 500–1,000. References, appendices and usually the abstract sit outside the count. If you are over the limit, the answer is cutting by structure, not shrinking every sentence.
“How long should a dissertation be?” is really three questions wearing one coat: what total has my university set, how should that total divide between chapters, and what do I do when my draft refuses to match either number. This guide answers all three with the figures UK students actually work to. It belongs to the chapter cluster alongside our guides to the abstract, the methodology and the conclusion — the four questions students ask at the start, middle and end of every project.
The totals, by level
Universities set the requirement, and the spread is wider than students expect. Undergraduate dissertations in the UK most commonly run from 8,000 to 12,000 words; some programmes, particularly in the humanities, set 7,500, and some science and engineering projects are shorter still because the artefact or experiment carries part of the credit. Masters dissertations most commonly run from 12,000 to 20,000; MBA and professional Masters projects often sit at the lower end, research Masters at the upper. PhD theses are a different world (typically 80,000 words or so) and outside this guide’s scope.
Three rules matter more than the headline figure. First, the count is a maximum or a target with a tolerance, and your handbook says which — most allow plus or minus 10 per cent without penalty, and penalise beyond it, so a 10,000-word dissertation at 10,900 words is usually safe and at 11,500 usually is not. Second, what counts is defined: the reference list is almost always excluded, appendices almost always excluded, the abstract usually excluded, tables and figure captions sometimes counted — check your handbook’s wording rather than assuming. Third, markers can tell: a padded literature review and a rushed discussion are visible in the proportions before a word is read, which is why the division matters as much as the total.
Dividing the total: a working split
For a standard empirical project of 10,000 words, the proportions that supervisors most often steer students towards are:
- Introduction — around 10 per cent (about 1,000 words): context, rationale, aim, research questions, and a map of what follows.
- Literature review — around 25–30 per cent (2,500–3,000 words): the largest single chapter in most projects, because it must build the whole case for your study. Our literature review help page covers how that chapter is structured.
- Methodology — around 15–20 per cent (1,500–2,000 words): philosophy, design, sampling, methods, analysis and ethics, each justified rather than merely described.
- Results or findings — around 15–20 per cent (1,500–2,000 words): findings presented in research-question order — our guide to writing the results chapter covers the format.
- Discussion — around 15–20 per cent (1,500–2,000 words): your findings interpreted against the literature — the analytical peak of the project.
- Conclusion — around 5–10 per cent (500–1,000 words): the research questions answered, contribution, limitations, recommendations.
Scale the percentages, not the habits, for a 15,000-word Masters project: the literature review and discussion usually grow fastest. Non-empirical projects (a purely theoretical or library-based dissertation) divide differently — no results chapter, a longer literature and analysis body — and humanities dissertations often run as three or four long chapters rather than the empirical skeleton above. Your supervisor’s steer outranks any general split, including this one.
Over the limit: cut by structure, not by nibbling
Almost every student who is over the limit tries to fix it by trimming sentences everywhere, loses two days, and removes 300 words. Structural cutting is faster and improves the grade as it goes:
- Find the chapter that broke its proportion. It is usually the literature review (sources summarised that should have been synthesised — cut the summaries, keep the synthesis) or the discussion repeating results (delete the repetition; the results chapter exists).
- Cut whole sub-sections that serve no research question. If a theme in the literature review never returns in the discussion, it is decoration — and decoration is what word limits exist to remove.
- Tighten tables and extracts. Qualitative findings chapters bloat on long extracts; keep the extracts that evidence the pattern, shorten the rest.
- Then, and only then, edit sentences. By this stage the remaining excess is usually small enough for an honest line edit — and if polishing is the remaining need, our proofreading and editing service works on your own words without rewriting them.
Going under the limit is rarer and usually signals a thin literature review or a discussion that reports rather than interprets; the fix is depth in those chapters, not filler elsewhere.
What about the abstract, appendices and the reference list?
These sit at the edges of the count and confuse students every year. The abstract — typically 200 to 350 words — is usually excluded from the word count but is always read, and read first; treat it as outside the arithmetic and inside the grade. Appendices are excluded and are for material the reader may consult (instruments, full tables, ethics approvals), never for analysis that belongs in the body; a marker is not obliged to read them, so anything your argument needs must live in the chapters. The reference list is excluded almost everywhere. Title pages, contents pages and declarations are administrative, not words. If your handbook is silent on any of these, ask your supervisor rather than guessing — the two-minute email is cheaper than the penalty.
A final note on planning backwards
The students who hit their number calmly are the ones who allocated it before drafting: chapter targets written down, drafts measured against them monthly, and the abstract — our abstract guide covers its 200–350 words — written last, when the project finally knows what it is. If a single chapter is the one breaking your total or your timetable, our dissertation chapter help service supports individual chapters modelled properly, so the rest of the project has a pattern to follow.
Dissertation length questions
Do references and appendices count towards the word count?
Usually the reference list is excluded and appendices sit outside the count, but the treatment of tables, figures, footnotes and in-text citations varies by university and sometimes by department. The rule that governs your project is the one in your course handbook — check it before you plan your chapters, not in submission week.
What happens if my dissertation is more than 10% over the limit?
Many universities allow a tolerance of around 10% and then penalise beyond it — some deduct marks in bands, others simply stop marking at the limit plus tolerance, so your conclusion may never be read. Because the sanction differs, the safe strategy is structural cutting: remove a weak section whole rather than nibbling sentences from every chapter.
Will a dissertation well under the word count lose marks automatically?
Not automatically — markers mark against criteria, not the counter. In practice, though, a project far under its target usually shows the gap somewhere the criteria do bite: thin literature coverage, underdeveloped analysis or a rushed discussion. If you are under and the work genuinely covers everything, padding will not help you; depth will.
How long should each chapter be in a 10,000-word dissertation?
A widely used split gives the introduction about 10%, the literature review 25%, methodology 15%, results 15%, discussion 25% and the conclusion 10%, with the abstract written separately at roughly 300 words. Treat those shares as a starting plan rather than a rule — empirical and theoretical projects weight their chapters differently.