To write a dissertation, work through six stages in order: choose a researchable topic and get a proposal approved, do the research (reading plus your own data), draft the chapters, then edit, format and submit before defending the work in a viva. Treating it as a sequence of small, deadline-bound deliverables — not one giant essay — is what gets students to a strong finish.
This overview orients you through the whole project and links to a detailed guide for every chapter, so you can zoom in wherever you are stuck without losing the bigger picture.
A dissertation is the longest single piece of writing most students ever produce, and it is built in stages rather than written in one sitting. The mistake that sinks most projects is treating it as one giant essay; the students who finish well treat it as a sequence of smaller, manageable deliverables, each with its own deadline. This guide walks you through the whole journey from a blank page to a bound submission, and links out to a detailed chapter guide at every step so you can go deeper wherever you need to.
Step 1: Choose a topic and write a proposal
Everything downstream depends on the question you choose, so do not rush this. A strong dissertation topic is narrow enough to answer in your word count, original enough to add something, and feasible given the data and time you actually have. Run a quick scan of recent journal articles in your area and look for a gap, a contradiction, or a method that has not yet been applied to your context. Our guide on how to find a good dissertation topic walks through this filtering process with worked criteria.
Once you have a question, you turn it into a proposal: the document that gets your supervisor and ethics committee to sign off before you spend months on the wrong thing. A proposal states your aim, objectives, a short literature rationale, and your intended method. The structure carries straight over from a research-degree pitch, so the framework in our PhD proposal structure guide applies to taught masters and undergraduate proposals too, including how to specify your methodology.
Step 2: Do the research
Research has two halves. First you read: you build a critical map of what is already known, which becomes your literature review. Use our guide on how to write a dissertation literature review to move from summarising sources to genuinely synthesising and critiquing them. As you read, decide which lens you are looking through, because your theoretical framework determines what counts as evidence and how you interpret it.
Second you collect and analyse your own data, whether that is interviews, a survey, an experiment, or a corpus of documents. If you are running a survey, design it carefully, because a weak instrument cannot be fixed later; our dissertation questionnaire guide covers question wording and structure. Keep meticulous notes on your method as you go, since you will need to reconstruct every decision when you write the methodology chapter.
Step 3: Write the chapters
Now you draft, and the order in which you draft matters. Most students write the methodology and results first (they are the most factual and the easiest to start), then the literature review, then the discussion, and they write the introduction and abstract last, once they actually know what the dissertation says. Before you write a word, fix the skeleton using our guide on how to structure a dissertation or thesis so every chapter has a clear job.
Each chapter is a craft in itself. Use the dedicated guides as you reach each one: writing the introduction, presenting your findings, writing the discussion, and writing the conclusion. Save the abstract for the very end.
| Stage | What you actually do | Typical timing (12-month project) |
|---|---|---|
| Topic & proposal | Scope a researchable question; write aim, objectives and method; get ethics approval | Months 1–2 |
| Literature review | Read widely, map the debate, identify your gap, draft the review | Months 2–4 |
| Methodology & data | Finalise method, collect data (interviews, survey, experiment), analyse it | Months 4–7 |
| Findings & discussion | Present results clearly, then interpret them against the literature | Months 7–9 |
| Introduction, conclusion, abstract | Frame the whole project and state its contribution, written last | Month 10 |
| Editing & formatting | Revise structure, proofread, format references, add front and back matter | Month 11 |
| Submission & viva | Hand in to the deadline; prepare to defend your choices | Month 12 |
Step 4: Edit, proofread and format
A first draft is not a dissertation; it is raw material. Edit in three passes: structural (does each chapter argue what it should and in the right order), line-level (clarity, signposting, academic tone), then proofreading (typos, grammar, consistent terminology). Leave 48 hours between writing and proofreading so you read what is on the page, not what you meant to write. Finally, build the front and back matter — title page, declaration, acknowledgements, contents, list of figures and tables, glossary and appendices — and lock your reference list to a single citation style.
- The research question is narrow, original and answerable in your word count
- Your supervisor and ethics committee have approved the proposal before data collection
- The literature review synthesises and critiques sources rather than listing them
- Every method choice is documented well enough to justify it months later
- Findings are reported neutrally; interpretation is kept in the discussion
- The introduction and abstract are written last, after the rest is done
- References follow one consistent style and every in-text citation appears in the list
- Front and back matter (title page, declaration, contents, appendices) are complete
- The whole document has been proofread after a break, not immediately after writing
A worked example: one topic through every step
From a vague interest to a defended dissertation.
Start (too broad): “I want to write about remote working.” That is a subject, not a question. After scanning recent HR journals, the student narrows it.
Refined question: “How does hybrid working affect the perceived team belonging of early-career employees in UK professional-services firms?” Narrow, contemporary, and answerable with a survey.
Proposal aim: “This study aims to examine the relationship between hybrid working arrangements and team-belonging perceptions among employees with under three years’ tenure.” Two objectives follow: (1) measure belonging across office-based, hybrid and remote staff; (2) identify which workplace practices moderate the effect.
Method: An online questionnaire (40 items) distributed to 220 employees, analysed with a one-way ANOVA and regression.
A findings sentence (neutral): “Mean belonging scores were significantly lower for fully remote employees (M = 3.1) than for hybrid employees (M = 4.2), F(2, 217) = 14.6, p < .001.”
A discussion sentence (interpretive): “This pattern supports Spreitzer’s social-isolation thesis and suggests that scheduled in-person days, rather than total office time, drive belonging — a practical lever for HR teams.”
Conclusion line: “The study’s contribution is to show that belonging is sensitive to the structure of contact, not its volume.” That single sentence is what the abstract, written last, then compresses for the reader.
Stuck on the dissertation, not the deadline?
Our UK-based academic writers and editors can help with any single chapter or the full project, topic to submission.
Frequently Asked Questions
Plan on three to nine months for a taught masters dissertation and two or more years for a doctoral thesis, depending on the data you collect. A useful split for a one-year project is roughly two months on the topic and proposal, two to three months on the literature review and data collection, two to three months on writing, and a final month on editing, formatting and viva preparation. Start the literature review while you wait for ethics approval so the early weeks are not wasted.
Word counts vary by level and institution, but typical ranges are 8,000–12,000 words for an undergraduate dissertation, 15,000–20,000 for a taught masters, and 70,000–100,000 for a PhD thesis. Always work to your department’s exact limit, which is usually stated as a maximum that excludes the abstract, references and appendices. Allocate words by chapter before you draft so no single section crowds out the others.
Start by narrowing a broad interest into one researchable question, then write the proposal that gets it approved. Once approved, do not begin with the introduction. Draft the methodology and findings first because they are the most factual, then the literature review, then the discussion, and write the introduction and abstract last. Fixing the chapter structure before you write keeps every section focused on its job.
The standard order is title page, abstract, acknowledgements, contents, introduction, literature review, methodology, findings or results, discussion, conclusion, references and appendices. That is the reading order, not the writing order — most students write the introduction and abstract last. Our guide on how to structure a dissertation or thesis sets out exactly what belongs in each chapter.
A good dissertation answers a clear, narrow question with evidence, and shows critical thinking rather than description. The hallmarks are a focused research question, a literature review that synthesises and critiques rather than summarises, a method described precisely enough to be repeated, findings reported neutrally, and a discussion that interprets those findings against existing work. Consistent referencing and clean formatting signal the same rigour to your examiner.
The findings (or results) chapter reports what you observed, neutrally and without interpretation — the numbers, themes or patterns your data produced. The discussion chapter then explains what those results mean: how they answer your research question, how they fit or challenge the literature, and what their limitations and implications are. Keeping the two separate is one of the most common marking criteria, so resist the urge to interpret while you are still reporting.
