The introduction is the chapter examiners read first and candidates write last, and the gap between those two facts explains most of what goes wrong with it. Written early, it becomes a promise the thesis does not keep. Written late but hurriedly, it becomes a summary of what you happened to do rather than an argument for why it needed doing. Written properly, it is the chapter that tells two busy academics, in twenty minutes, that they are reading a doctorate.
What follows is the chapter as seven moves, in the order to draft them. Each has a stated output, so you can tell whether you have finished it. Research degree regulations differ by institution — some prescribe an introduction structure, most do not — so read your own code of practice and any departmental thesis guidance before you commit to a shape.
Before anything: write it last, and know why
An introduction has to promise exactly what the thesis delivers. Until the findings are settled and the discussion has been drafted, you do not yet know precisely what that is — which is why an introduction written in year one is always rewritten, and why drafting it before the discussion chapter means writing it twice.
There is one honourable exception. Most candidates need a working introduction earlier for the upgrade or confirmation review, and that is a legitimate use — but treat it as a scaffold with a demolition date, not a draft. When you return to it in the final months, rebuild rather than edit. The whole method for converting late-stage material into finished chapters is in our notes-to-chapter workflow.
Output of this step: a decision, recorded in your writing plan, that the introduction is drafted after the discussion — and a note of the one week you have allocated to it.
Move 1: Establish the territory
Open on the field, not on your project. The reader needs to know what area of research this is, why it is an active area, and what is generally accepted within it, before any of your claims can land.
The classic description of this shape is John Swales’s CARS model — Create A Research Space — set out in Genre Analysis (Cambridge University Press, 1990), which observed that research introductions across disciplines tend to move through establishing a territory, establishing a niche within it, and then occupying that niche. It is a description of what competent introductions do, not a rule imposed on you, and it scales from a journal article’s opening paragraphs to a doctoral chapter of several thousand words.
Two failure modes to avoid here. Starting too wide — a doctorate in nurse retention does not begin with the founding of the NHS. And starting inside your own project, which leaves the reader unable to judge whether anything you say next matters.
Output: three to six paragraphs that a competent researcher from a neighbouring field could read and then correctly state what area you work in and why it is contested.
Move 2: Establish the gap — and be careful how you do it
This is where most introductions are weakest, and the weakness has a name in the literature. Reviewing 52 articles in organization studies, Sandberg and Alvesson found that the most common way of constructing a research question “across paradigmatic camps is to spot various ‘gaps’ in the literature” — and called that dominance surprising, “given it is increasingly recognized that theory is made interesting and influential when it challenges assumptions that underlie existing literature” (Organization, 2011, 18(1), 23–44). Their companion paper in the Academy of Management Review (2011, 36(2), 247–271) proposes problematisation — surfacing and challenging the assumptions a literature rests on — as the stronger alternative.
The practical consequence for your chapter is not that gaps are forbidden. It is that “nobody has studied X in Y” is the weakest available justification, because the obvious reply is that nobody needed to. Stronger moves, in ascending order of force:
- An under-examined case with stated consequences. Not “no one has studied this in Wales”, but why the Welsh case should behave differently, and what follows if it does.
- A contradiction in the evidence. Two well-conducted bodies of work disagree, and nobody has explained why.
- A methodological limitation shared across the field. The finding may be an artefact of how everyone has measured it.
- A challenged assumption. The literature takes something for granted; you show it may not hold, and that a great deal follows from that.
Output: one paragraph naming the problem so specifically that a reader could go to the literature and check that you are right. Vagueness here is the single most reliable predictor of a weak thesis.
Move 3: State the aim, the questions and the objectives — and make them agree
Three different things, routinely conflated, and examiners notice when they drift apart.
- The aim is what the thesis is for, in one sentence. “This thesis aims to explain why…”
- The research questions are what you actually answer — usually two to four. More than four is a warning sign that the project has not been narrowed.
- The objectives are what you do in order to answer them, and they map onto chapters.
Then run the consistency check, which takes twenty minutes and saves a correction: every question must be answered somewhere in your findings, every finding must belong to a question, and the methods chapter must describe a design capable of answering each one. Drift between these three lists is one of the most common corrections examiners impose, precisely because it is visible without close reading.
Output: a numbered list of questions, and a table for your own use mapping each question to the chapter that answers it.
Move 4: State the contribution in one sentence
This is the sentence the entire examination circles, and the one candidates most often leave implicit in the hope that the thesis will speak for itself. It will not. Write it explicitly, in the introduction, in a form a reader could quote back to you.
What makes a contribution statement work is that it names what is now known that was not known before — not what you did. Compare:
Weak: “This thesis contributes to the literature on supervision by exploring the experiences of doctoral researchers in three UK institutions.”
Strong: “This thesis shows that doctoral researchers evaluate supervisory availability against disciplinary norms of independence rather than against contact hours, which explains why identical supervision arrangements produce opposite satisfaction judgements across departments — a mechanism absent from existing accounts, which treat availability as a quantity.”
The weak version describes an activity. The strong version makes a claim, states the mechanism, and says what it displaces. It is also, not coincidentally, an answer to the first question of most vivas. What counts as sufficient originality — and the forms it can take — is set out in our guide to the original contribution to knowledge.
Output: one sentence you can say aloud without reading it.
Move 5: Draw the boundaries
Say what the thesis does not do, and why. A short scope section — one or two paragraphs — pre-empts a whole category of examiner question by showing that the omission was a decision rather than an oversight.
Cover the population or setting you are not addressing, the theoretical traditions you are not engaging, and the methodological roads not taken. Keep it to genuine boundary decisions: this is not the place for the limitations of your findings, which belong with the findings themselves and again in the discussion.
Output: a paragraph beginning, in substance if not in wording, “This thesis does not attempt to…”.
Move 6: Map the thesis
Close with a chapter-by-chapter outline: one short paragraph per chapter saying what it does and what it establishes for the next one. Written well, this is not filler — it is the reader’s proof that the thesis has an argumentative sequence rather than a sequence of tasks.
The test of a good map is that each entry states a function, not a topic. “Chapter 4 presents the interview findings” is a table of contents in prose. “Chapter 4 establishes the three interpretive frames participants used, which the analysis in Chapter 5 then tests against the survey data” is a map.
Output: one paragraph per chapter, each naming what that chapter establishes.
Move 7: Write the opening paragraph last of all
The first paragraph of the introduction is the first paragraph of the thesis, and it is the hardest hundred words in the document. Write it once everything else exists, when you know exactly what you are introducing.
It needs to do only two things: give the reader the subject, and give them a reason to keep reading. A concrete opening — a number, a case, a contradiction — outperforms an abstract one almost every time.
How long should it be, and where does the literature review go?
Conventions vary by discipline and institution, and your handbook governs. As a working guide, introductions in monograph-style theses commonly run between about 5 and 10 per cent of the total — five to eight thousand words in an eighty-thousand-word thesis — with the shorter end typical in the sciences and the longer end in the humanities and social sciences.
On the literature question, there are two established patterns and you should pick one deliberately. Either the introduction gives a compact positioning of the field and a separate chapter does the full review, which is the most common shape; or the introduction and literature review are integrated into one long opening chapter, which is normal in some humanities disciplines and in theses by publication, where the framing chapter carries the positioning for several papers at once. What fails is doing half of each: a review too long to be positioning and too thin to be a review.
The three failures examiners name
- The promise the thesis does not keep. An introduction written early and never revised, announcing four research questions when the thesis answers two. This is the argument for writing it last, and it produces corrections that are tedious rather than difficult.
- The contribution left implicit. The work is original and the chapter never says so. Examiners are not obliged to construct your claim for you, and in the viva you will be asked to state it anyway — the eight questions almost every candidate faces are set out in our guide to viva preparation, and the contribution question is the one the rest hang from.
- The literature review that ate the introduction. Thirty pages of review before the reader has been told what the thesis argues. Position, then promise, then review.
Drafting it without losing the week
The introduction is a structured writing task with a known shape and a fixed set of outputs, which makes it the wrong chapter to write into a blank document. Build the seven moves as a skeleton first, drop your material under each one, and draft against that rather than against the whole thesis at once.
Tesify structures and drafts the chapter with you — the move-by-move architecture, the drafting workspace and an automatic bibliography that keeps your references formatted while you write. The territory, the problem, the contribution and every sentence claiming them stay yours: 100% written by you.
Frequently asked questions
How long should a PhD thesis introduction be?
Commonly around 5–10 per cent of the thesis — roughly five to eight thousand words in an eighty-thousand-word monograph — but conventions differ sharply by discipline and your institution’s guidance governs. Length matters far less than whether the chapter completes all seven moves.
Should I write the introduction first or last?
Last, after the discussion. The chapter has to promise exactly what the thesis delivers, and you do not know that precisely until the findings and their interpretation are settled. A working version for your upgrade review is a scaffold, not a draft.
What is the difference between the introduction and the literature review?
The introduction positions the thesis and states its claim; the literature review evaluates the body of work the claim engages. Some disciplines merge them into one opening chapter, which is fine if done deliberately — the failure is a hybrid that is too long to position and too short to review.
Do I need to state my contribution in the introduction?
Yes, explicitly and in one sentence. Leaving it implicit does not protect you from the question; it just means the examiners ask it in the viva without a prepared answer on the page.
How many research questions should a PhD have?
Typically two to four. More usually indicates a project that was never narrowed, and each additional question is another thing your methods must be capable of answering and your findings must actually address.
Should the introduction include definitions?
Define only the terms the reader needs in order to follow the chapter, and do it briefly. Full conceptual definitions belong in the literature review or theoretical framework, where you can argue for them rather than assert them.
Is it acceptable to say “this thesis argues” in the first person?
In most UK disciplines now, yes — first person is increasingly accepted and in reflexive qualitative work it is expected. Check your department’s convention and, more importantly, a recent successful thesis from your own school.
What if my research questions changed during the PhD?
Entirely normal, and the introduction is where you present the final ones as though they had always been the plan. Any narrative of how the project evolved belongs in the methodology, if anywhere.
Should the introduction mention my methodology?
A sentence or two naming the approach and the data, so the reader knows what kind of study this is. The justification belongs in the methods chapter and repeating it here dilutes both.
Do examiners actually read the introduction closely?
Closely, and first. It is where they form the working hypothesis about your thesis that the rest of the reading tests, which is why the contribution sentence is worth a disproportionate share of your final week.
