How to Write a Thesis: Complete Step-by-Step Guide for 2026

Most students know they need to write a thesis. Very few know how to actually start one. The blank document is intimidating, the deadline feels distant (until it does not), and the sheer scope of the task — original research, structured argument, formal presentation — can trigger weeks of productive-looking avoidance. This guide cuts through that paralysis. It shows you exactly how to write a thesis, chapter by chapter, from the first day you open your research notebook to the morning you submit.

The advice here draws on the thesis-writing traditions of Oxford, Cambridge, MIT, Harvard, and Stanford — institutions where the thesis is treated as a foundational scholarly act, not a hurdle. Whether you are writing a 15,000-word undergraduate dissertation or an 80,000-word PhD thesis, the structural logic is the same. Adapt the scale; keep the method.

Step 1: Choose a Researchable Topic

The single biggest mistake thesis writers make is choosing a topic that is too broad. “Climate change and society” is a PhD programme, not a thesis. “The effect of carbon pricing schemes on household energy consumption in Germany, 2015–2025” is a thesis topic. A good topic has three properties:

• Scope: It can be answered in the word count you have been given.
• Researchability: Data, literature, or participants exist that you can actually access.
• Contribution: It adds something — even a small, specific thing — to existing knowledge.

The best topic selection method is a gap analysis. Read 20–30 recent papers in your area. Look for phrases like “future research could examine…”, “limited work has addressed…”, and “a longitudinal study would be valuable…” — these are invitations. Your thesis answers one of them.

For deeper guidance on narrowing your focus, see our article on how to choose a thesis topic.

Step 2: Write Your Thesis Proposal

A thesis proposal is a contract between you and your supervisor. It is typically 1,000–3,000 words and covers: background and rationale, research question(s), preliminary literature review, proposed methodology, timeline, and expected contribution. Some programmes require a formal proposal defence.

The proposal forces clarity early. If you cannot write a clear rationale in 500 words, you do not yet understand why your research matters. Use the proposal process to test your idea: share it with your supervisor, a peer, and ideally someone outside your discipline. If they cannot understand the value of your work, refine the framing.

Oxford and Cambridge supervisors often require three rounds of proposal revision before approval. That iterative friction is not bureaucracy — it is the mechanism that produces focused, defensible research.

Step 3: Conduct Your Literature Review

The literature review does two jobs: it proves you know the field, and it builds the logical case for your research question. It is not a summary of everything ever written on the topic. It is a critical, thematic analysis that identifies what is known, what is contested, and what is missing.

A systematic approach works best:

1. Define your search terms and databases (Google Scholar, JSTOR, Web of Science, PubMed for STEM).
2. Set inclusion/exclusion criteria (date range, language, publication type).
3. Screen titles and abstracts, then read full texts for relevant papers.
4. Use a reference manager (Zotero, Mendeley) to organise your sources.
5. Code papers thematically: identify recurring arguments, conflicting findings, and methodological patterns.
6. Write the review around themes, not around individual papers.

A strong literature review at master’s level synthesises 40–80 sources. At PhD level, 100–200 is common. Quality matters more than quantity — one well-analysed study is worth five summaries.

See our full guide on literature review structure with annotated examples for practical templates.

Step 4: Design Your Methodology

Your methodology chapter explains how you answered your research question — and justifies why you chose that approach over available alternatives. It has three layers:

• Research philosophy: Are you a positivist (objective reality, measurable) or interpretivist (subjective meaning, context-dependent)? This shapes everything else.
• Research approach: Inductive (theory from data) or deductive (testing existing theory)?
• Research methods: Quantitative (surveys, experiments, statistical analysis), qualitative (interviews, ethnography, discourse analysis), or mixed methods?

Whatever you choose, justify it with reference to your research question. A PhD candidate at MIT told us her examiners spent 40 minutes on her methodology chapter at the viva — twice the time spent on her results. Methodology is where you demonstrate scholarly rigour.

Step 5: Write the Body Chapters

Most theses have between three and five body chapters after the literature review and methodology. For empirical theses, the standard structure is: Results / Findings → Discussion. For theoretical or humanities theses, chapters typically develop stages of a central argument.

Practical writing advice:

• Write chapter summaries before you write the chapters. A 200-word summary per chapter takes one hour and saves weeks of structural revision.
• Start each chapter with a signpost paragraph: what this chapter does and how it connects to the overall argument.
• End each chapter with a brief conclusion that links forward to the next.
• Write every day, even 200 words. A Harvard study on academic writing productivity found that daily writers produce 3–5 times more output than binge writers.
• Do not edit while you draft. First drafts exist to get ideas on the page. Editing is a separate, later stage.

Tools like Tesify Write can help you draft and refine individual sections, check for structural gaps, and generate suggestions when you are stuck — without writing your thesis for you. The distinction matters: AI assistance is a legitimate productivity tool when you remain the scholar making analytical decisions.

Step 6: Write Introduction and Conclusion Last

This surprises many students: the introduction and conclusion are the last things you write, not the first. Here is why: until your body chapters exist, you do not know precisely what you are introducing or concluding.

A strong thesis introduction covers:

• The context and significance of your topic (why it matters now)
• The research gap your work addresses
• Your research question(s) and objectives
• Your theoretical framework (briefly)
• An overview of your methodology
• A chapter-by-chapter roadmap

The conclusion mirrors the introduction. It restates your research question, summarises your key findings, discusses their implications, acknowledges limitations, and recommends future research. A Stanford writing guide describes the conclusion as “the introduction’s promise, kept” — a useful mental model.

For full examples, see our guide on thesis introduction structure with real examples.

Thesis Structure at a Glance

Step 7: Revise, Proofread, and Submit

A first draft is never a submission draft. Professional academic writers revise three to five times before submitting. The revision process has distinct stages:

1. Structural revision: Does the argument flow logically? Does each chapter earn its place? Does the conclusion answer the research question stated in the introduction?
2. Argument revision: Are your claims supported by evidence? Are counter-arguments addressed? Is your contribution to knowledge clear?
3. Language revision: Is every sentence clear and concise? Are academic conventions (passive vs. active voice, hedging language) followed consistently?
4. Citation revision: Does every in-text citation have a matching reference? Are all references formatted correctly in your required style (APA, Harvard, MLA)?
5. Plagiarism check: Run your final draft through a plagiarism checker before submission. Even unintentional duplication of phrasing can be flagged as academic misconduct. Tesify’s Plagiarism Checker scans against academic databases and provides a similarity report with source-level detail.

Submit at least 48 hours before the deadline. Last-minute technical issues — file format errors, system outages, corrupted documents — are not accepted as extenuating circumstances at most institutions.

For writing guidance in other languages, the same structured approach applies: French-language resources are available at Tesify.fr, Spanish at Tesify.es, and German at Tesify.io.

Frequently Asked Questions