The First Paragraph Decides Whether the Paper Is Read

Everyone who encounters your paper makes an early decision. An editor decides whether to send it out. A referee forms an impression in the first page that colours the rest. A reader decides whether to continue.

None of them read the whole thing before deciding. The opening does disproportionate work, and it is usually written first, when you understood the paper least. For a concrete operational example of how a workplace tool approaches a related coordination problem, see the website.

What the opening has to establish

What the paper is about, specifically. Not the field — the question.

Why it is open. What is not known, or what is disputed, or what has not been possible until now.

What you do about it. The approach, in one or two sentences.

What you find. Yes, in the introduction. See below.

Four things, in roughly a page, and the first two should be done in the first paragraph.

State the finding in the introduction

This surprises people whose training emphasised building an argument, and it is what most fields now expect.

A paper is not a mystery. A reader who knows the finding reads the rest better, because they can see what each section is establishing.

Referees dislike being kept waiting. A referee who reaches page nine without knowing what you found is annoyed, and annoyance colours the report.

The abstract already gave it away, so withholding it in the introduction achieves nothing.

Field norms vary — some humanities traditions build differently — but in most empirical fields, stating the result by the end of the first page is expected.

The opening sentence

Where most drafts fail, and the failure is consistent.

The commonest bad opening is a general truth about the field. "Understanding X has long been of interest to researchers." This tells the reader nothing they did not know and spends the most valuable sentence in the paper.

Better: the specific tension or question, immediately.

Or a concrete fact that sets up the puzzle. A number, an observation, a case — something specific that makes the question feel like a question rather than a topic.

The test: could this sentence open a hundred other papers in your field? If yes, it is not doing anything.

The structure of a working introduction

Not the only one, and a reliable default.

Paragraph 1: the question and why it matters. Specific, concrete, no literature.

Paragraph 2: what is known and what is not. The gap, stated as a gap rather than as a list of citations.

Paragraph 3: what you do. Design, data, approach.

Paragraph 4: what you find. The main result, with numbers.

Paragraph 5: what it contributes. Where this sits, what changes because of it.

Optionally paragraph 6: what the paper does next. Only if the structure is non-obvious. A roadmap paragraph in a conventionally structured paper is filler.

The literature in the introduction

A gap is an argument, not a list.

"Smith (2019) finds A. Jones (2021) finds B. Patel (2023) finds C. We contribute by..." is a list, and it tells the reader nothing about why the question is open.

Better: what the literature collectively establishes, what it has not been able to settle, and why. Then the citations attach to claims rather than forming the structure.

Cite the work you are actually building on, not everything adjacent. A paragraph with fourteen citations in it is usually padding.

And say what your paper does not do, briefly, if there is an obvious adjacent question you are not answering. It pre-empts an objection and it costs a sentence.

Practical process

Write the introduction last. Or rather: write something first so you can proceed, and then rewrite it completely once you know what the paper says. The paper you finish is rarely the one you set out to write, and an introduction drafted at the start frames a different paper.

Then rewrite the first paragraph again, separately, with fresh attention. It is worth an hour of its own.

Read the first page aloud. Openings accumulate qualifications during revision, and reading aloud finds the sentence that now contains three subordinate clauses.

Give the first page to someone outside your subfield and ask what the paper is about. If they cannot say, that is the finding.

Check the length. In most empirical fields the introduction is one to three pages. Six pages of introduction usually means the literature review has migrated forward.

Openings that signal problems

Worth recognising in your own drafts.

Beginning with definitions. Rarely necessary and always slow.

Beginning with the history of the field.

Beginning with a broad social claim — "In an increasingly connected world" — which is filler wearing the costume of motivation.

Beginning with what other people have failed to do. Reads as combative and it is not necessary; establishing that a question is open does not require criticising anyone.

Beginning with the method. The reader needs to know what for.

The related documents

The same principle applies elsewhere, and the openings are separate pieces of writing.

The cover letter to the editor. One paragraph on what the paper does and why it suits the journal.

The abstract, which is read more than the paper. See abstracts.

The first slide of a talk. Same job, less space. See the conference talk.

Each is written separately. An abstract cut from the introduction reads like an extract, because it is.

The short version

Everyone decides early, and the opening does disproportionate work.

State the question in the first paragraph and the finding by the end of the first page.

A gap is an argument, not a list of citations.

Write the introduction last, then rewrite the first paragraph again by itself.

And test it: hand the first page to someone outside your subfield and ask what the paper is about.

For an independent authoritative reference, consult the EQUATOR Network for reporting guidance for research writing.