How to Start a Thesis: Proposal to Draft

“Starting” a thesis is usually a planning and approval problem before it is a writing problem. Most of the paralysis that shows up as “I don’t know how to start” actually comes from three unresolved decisions: the research question isn’t narrow enough yet, the proposal hasn’t been formally approved, or the writer is trying to draft the introduction before knowing what the rest of the document will argue. This guide walks through the sequence that actually gets a thesis moving — narrowing the topic, getting the proposal approved, building a working structure, and choosing where to write first — rather than generic writing-motivation advice.

A note on terminology: this guide uses “thesis” loosely to mean the long-form original-research document required for a graduate degree, whether your institution calls it a thesis or a dissertation. The two terms map differently by country and by degree level — see CASRAI’s Dissertation vs. Thesis comparison and the What Is a Dissertation? guide for the structural and terminological detail. Check your own graduate school’s style guide for which term and which structural requirements apply to you specifically — this guide covers principles that hold across most institutions and disciplines, not any one program’s exact template.

Before you draft anything: the proposal stage

In most graduate programs, “starting” the thesis formally means getting a proposal (sometimes called a prospectus, or in some UK/Commonwealth programs a “confirmation of candidature” document) approved by an advisor and, often, a committee. This step exists before any substantial drafting because it is where scope gets tested — a committee’s job at this stage is largely to stop a project that is too broad, too vague, or not yet feasible from consuming months of research effort before anyone catches the problem.

Narrowing the research question

A workable thesis question is narrow enough to be answered with the time, access, and methods actually available to a single graduate researcher, and specific enough that a reader can tell what evidence would confirm or disconfirm it. “Broad topic” and “research question” are not the same thing: a topic (“climate policy in coastal cities”) becomes a question only once it specifies a population, a mechanism, a comparison, or a gap in the existing literature that the thesis will address. Most advisors will push back on a proposal until this narrowing has actually happened — treat that pushback as the expected first round of the process, not a sign the project is failing.

Advisor and committee formation

Formal committee composition rules vary by institution and degree level, so check your graduate school’s own requirements rather than assuming a fixed committee size or membership rule. What is consistent across most programs is that the advisor relationship and, once formed, the committee’s expectations should be clarified in writing early — meeting cadence, who reviews drafts before the advisor does, and what the committee expects to see at the proposal stage versus later chapter reviews.

Ethics and regulatory approval timing

If the thesis involves human subjects, animal subjects, or other regulated research activity, institutional ethics approval (an Institutional Review Board or equivalent) is typically required before data collection begins — not before the proposal is written, but before any fieldwork, interviews, surveys, or specimen collection starts. Build the approval timeline into your project plan at the proposal stage: IRB review can take weeks to months depending on risk level and institution, and starting that process late is one of the most common causes of an otherwise well-planned thesis timeline slipping.

Build a working structure before you write full prose

Once the proposal is approved, the next step that actually unblocks writing is a structural outline — a chapter-by-chapter skeleton with a one- or two-sentence statement of what each chapter is responsible for. This does two things: it turns “write the thesis” (an intimidating, undifferentiated task) into a sequence of smaller, addressable ones, and it gives your advisor something concrete to react to before you’ve invested weeks in prose.

The specific chapter structure varies substantially by discipline. Empirical work in the sciences and social sciences generally follows some variant of the IMRaD structure (Introduction, Methods, Results, and Discussion) stretched across full chapters rather than journal-article-length sections, often with a dedicated literature-review chapter inserted after the introduction. Humanities theses are more often organized thematically or chronologically around a central argument, without a discrete methods chapter. Neither is more correct than the other — the structure should follow the discipline’s own conventions and the argument you’re actually making, not a template imported from a different field.

A common skeleton for empirical work:

Introduction — states the research question, its significance, and the thesis’s contribution
Literature review — synthesizes existing work and identifies the specific gap the thesis addresses
Methodology — describes and justifies the research design, data, and analytical approach
Results/findings — reports what the research produced, without interpretation
Discussion — interprets the findings against the literature and the original question
Conclusion — states the contribution, limitations, and implications

Where to actually start writing

Most graduate researchers do not write a thesis in chapter order, and the introduction is usually the wrong place to start despite being chapter one. An introduction needs to state the thesis’s contribution and preview its argument — both of which are genuinely hard to write accurately before the rest of the document exists to preview. A more common and lower-friction sequence:

Literature review first. You’ve likely already done much of this reading for the proposal. Turning proposal-stage reading notes into a structured review — see CASRAI’s how to write a literature review guide for building a synthesis rather than a source-by-source summary — produces early momentum and clarifies exactly what gap the thesis fills, which then makes the introduction easier to write later.
Methodology next, particularly for empirical work — this section is often the most mechanical to draft because it describes decisions you’ve already made and can typically be written accurately before data collection or analysis is complete.
Results/findings and discussion once analysis is underway or complete.
Introduction and conclusion last, or at minimum revised substantially at the end — both chapters make claims about the whole document, so they’re most accurate once the whole document actually exists.

This isn’t a universal rule — some writers genuinely think better by drafting a rough introduction first as a planning device, then rewriting it entirely at the end. The point isn’t the specific order; it’s that treating “start the thesis” as synonymous with “write chapter one first” is what generates most of the stalling. Pick whichever entry point produces actual pages, and expect to substantially revise the introduction regardless of when you first draft it.

Setting a realistic timeline

A thesis timeline that only has one milestone (“submit draft to advisor”) invites the same paralysis as an unstructured writing task. Break the timeline into chapter- or section-level targets with real dates, and share them with your advisor so there’s an external checkpoint rather than only a private deadline. For empirical theses, build the ethics-approval and data-collection timeline in explicitly rather than treating it as a background assumption — approval delays are one of the most common reasons a thesis timeline needs to be renegotiated partway through.

Common pitfalls when starting

Scope that’s still topic-sized, not question-sized. If the proposal can’t state in one or two sentences what specific question the thesis answers, it usually isn’t ready to move into drafting yet.
Treating the literature review as a summary rather than a synthesis. A chapter that walks through sources one at a time, rather than organizing them around themes, debates, or a gap, reads as a reading list rather than an argument — and is one of the most common reasons a full draft comes back from an advisor for a substantial rewrite.
Waiting for IRB or equivalent approval to start writing anything. Approval gates data collection, not the literature review, methodology justification, or introduction framing — those can generally proceed in parallel with the review process.
Drafting the introduction as if it were an essay hook. A thesis introduction is a structural document — it states the question, the contribution, and the chapter roadmap — not a rhetorical opening designed to draw a general reader in the way a magazine article might.

Frequently asked questions

Do I need IRB or ethics approval before I start writing my thesis?

No — ethics approval is required before you begin data collection involving human or animal subjects, not before you start writing. The literature review, methodology justification, and introduction framing can typically be drafted while approval is in process, though you should not begin fieldwork, interviews, surveys, or specimen collection until approval is granted.

Should I write the introduction chapter first?

Usually not, even though it’s chapter one in the final document. Most writers find the literature review or methodology chapter easier to draft first, since the introduction needs to accurately preview a document that doesn’t fully exist yet. It’s common, and expected, to substantially rewrite the introduction after the rest of the thesis is drafted.

What’s the difference between a thesis proposal and the thesis itself?

A proposal (sometimes called a prospectus) is a shorter planning document — typically covering the research question, its significance, a review of relevant literature, and the proposed methodology — submitted for advisor or committee approval before the bulk of research and writing begins. The thesis itself is the full document reporting the completed research. Length and formality requirements for the proposal vary substantially by institution and degree level.

Is “thesis” or “dissertation” the correct term for my degree?

It depends on your country and institution. In the US and Canada, “dissertation” typically refers to the doctoral-level document and “thesis” to a master’s-level one; in the UK, Ireland, Australia, and much of Europe this is usually reversed. See CASRAI’s Dissertation vs. Thesis comparison for the full breakdown, and confirm against your own graduate school’s terminology since individual institutions sometimes deviate from their country’s general convention.

How to Start a Thesis: Proposal, Structure, and the First Chapter