Structuring a Paper Introduction: The Funnel

A research paper’s introduction has a conventional shape: it opens on the broad field, narrows step by step to the specific problem the paper addresses, names the gap in existing knowledge, and closes by stating what the study actually did about it. Writing teachers and journal style guides call this the funnel structure (sometimes the “hourglass” model, since a strong paper widens back out again in the Discussion). Getting the funnel right is a distinct skill from knowing that a paper follows IMRaD structure overall, or knowing how to write the separate literature review that a thesis or grant proposal might require — this guide is specifically about how to organize the four or five paragraphs that make up the Introduction section itself.

Why the introduction follows a funnel shape

A reader — and, just as practically, a peer reviewer skimming for the paper’s contribution before deciding whether to read closely — needs three things established in order before the rest of the paper makes sense: what field this sits in, what specifically is unresolved, and what this particular study contributes to resolving it. Starting with narrow technical detail before the reader has that orientation is a common reason introductions read as confusing even when every individual sentence is accurate.

The funnel logic is formalized in composition pedagogy as the CARS model (“Create a Research Space”), developed by linguist John Swales in his 1990 book Genre Analysis: English in Academic and Research Settings from analysis of how published research-article introductions actually move through their argument. CARS describes three “moves,” each with recognized sub-steps:

Move 1 — Establishing a territory: claiming centrality of the topic, making topic generalizations, and/or reviewing prior research.
Move 2 — Establishing a niche: counter-claiming, indicating a gap, raising a question, or continuing a tradition — i.e., stating what is not yet known or resolved.
Move 3 — Occupying the niche: outlining purposes, announcing present research, announcing principal findings, and/or indicating the paper’s structure.

Swales’ three moves map closely onto the four practical steps most journals and supervisors expect, which this guide covers in order below: broad context, narrowing to the problem, the gap, and the aim/hypothesis statement.

Step 1: Establish the broad field context

Open with why the general topic matters — to the field, to practice, or to a real-world outcome — not with a narrow technical claim the reader has no context for yet. This is typically one to three sentences, not a paragraph of throat-clearing. Concretely, this step usually does one of:

States the scale or significance of the general phenomenon (a disease burden, an unresolved theoretical question, a widely used but poorly characterized method).
Orients a reader outside the immediate subfield to why the topic is studied at all.
Cites one or two landmark or highly cited sources establishing the field, not an exhaustive review — that belongs in a dedicated literature review, not the paper’s own introduction.

A common failure mode here is starting too broad (“Cancer is a leading cause of death worldwide…”) in a way that could preface almost any oncology paper — it establishes no real territory and wastes the reader’s limited attention before the funnel has narrowed at all. The territory claim should already be recognizably adjacent to this specific paper’s subfield.

Step 2: Narrow to the specific problem area

The middle of the funnel moves from the general field to the specific line of work this paper sits within — the particular mechanism, population, method, or theoretical question, and what is currently understood about it. This is where the introduction does its compressed literature-framing work: not a full review, but enough prior-work context that a reader can see what’s already established and where this paper’s specific corner of the field begins.

This differs in scope and purpose from a standalone literature review: a literature review’s job is to comprehensively map and synthesize a body of work (often as its own chapter, section, or paper). An introduction’s narrowing paragraphs cite only what’s necessary to justify that a gap exists and that this study is a reasonable way to address it — typically a handful of the most directly relevant, recent studies rather than the field’s full history.

Step 3: Identify the gap

This is the pivot of the whole introduction, and the step most often written too weakly. The gap statement has to do real argumentative work: it must follow logically from the narrowing in Step 2 (a gap that doesn’t connect to the prior paragraph reads as unmotivated), and it must be specific enough that a reader can see exactly what is missing — not simply that “more research is needed.”

Gap statements typically take one of a few recognizable forms, corresponding to Swales’ “establishing a niche” move:

Knowledge gap: the specific question has not been directly studied, or not in this population/context/method.
Contradiction gap: existing studies disagree, and this work is designed to help resolve the discrepancy.
Methodological gap: prior work exists but used a method (design, instrument, sample) with a known limitation this study addresses.
Extension gap: prior work established a finding in one context, and it is unknown whether it generalizes to another.

A gap statement written as a vague appeal (“despite this, few studies have examined…”) without naming what specifically is missing, and why it matters, is a frequent reviewer criticism — vague gap statements read as a justification invented after the fact rather than the actual motivation for the study design.

Step 4: State the aim, objective, or hypothesis

The introduction closes by naming exactly what the study did in response to the gap just identified — the funnel’s narrowest point. Conventions differ slightly by field and journal, but this closing statement is typically one of:

A stated aim or objective (“This study examined whether…”, “The objective of this study was to…”), common across most empirical fields.
An explicit hypothesis (a predicted, testable relationship between variables), expected in fields and study designs where a directional or null prediction is standard practice (much experimental and quantitative research).
A brief statement of approach alongside the aim (“…using a randomized, cross-over design”), when naming the method briefly here helps orient the reader before the full Methods section.

This statement should map directly onto what the paper’s Methods and Results sections actually deliver — an aim promising more than the study design can support (or a hypothesis the study wasn’t actually powered or designed to test) is a mismatch reviewers catch quickly. Where sample size or statistical power is part of what justifies the study’s ability to test its stated aim, that justification itself belongs in the Methods section, not the introduction — see Justifying Sample Size in a Manuscript for how that argument is built separately.

How long should an introduction be?

There is no single universal word count — journal instructions-for-authors vary, and structured formats (e.g., a “Background” subsection under a journal’s own template) can prescribe different limits. As a general, non-binding convention seen widely in research-article introductions: three to five paragraphs, each doing one clear job in the funnel above, is typical for a standard original-research article — noticeably shorter than a thesis introduction chapter, which typically must justify the whole study at greater length and often incorporates a fuller literature review directly. Always check the target journal’s specific instructions for authors rather than assuming a fixed length.

Common pitfalls in introduction structure

Starting too broad, too long: spending an entire paragraph on field-level context that could preface any paper in the discipline, delaying the actual narrowing.
No clear pivot into the gap: the narrowing paragraphs and the gap statement read as disconnected, so the gap feels asserted rather than demonstrated.
Gap stated too vaguely: “little is known about X” without saying what specifically is unknown or why it matters enough to study.
Aim doesn’t match the gap: the closing aim/hypothesis addresses something slightly different from the gap just described, breaking the funnel’s logical chain.
Front-loading results or literature-review depth: pulling material that belongs in Results or in a dedicated literature review into the introduction, overloading it past its actual job.
No aim/hypothesis statement at all: ending the gap paragraph without ever explicitly stating what the study did, leaving the reader to infer the paper’s purpose from the Methods section instead.

How this relates to the rest of the paper

The funnel structure is the internal logic of one section within the paper’s larger IMRaD structure (Introduction, Methods, Results, and Discussion) — ICMJE describes IMRaD as “not an arbitrary publication format but a reflection of the process of scientific discovery,” and the introduction’s funnel is that same logic applied at the section level: context, then problem, then what this study specifically did about it. The Discussion section conventionally reverses the funnel — moving back out from specific findings to their broader significance — which is why the paired shape is sometimes called an hourglass; see Writing the Discussion Section of a Research Paper for that structure. For guidance on the separate task of comprehensively surveying prior work (as its own chapter or section, distinct from an introduction’s brief narrowing), see How to Write a Literature Review for a Research Paper. For general manuscript formatting conventions beyond section structure, see Research Paper Format: Margins, Fonts, Spacing, and Style Rules.

Frequently asked questions

Is the CARS model required by journals?

No — CARS is a descriptive academic-writing framework, not a formal publication standard like ICMJE’s recommendations or a journal’s own instructions for authors. It is widely taught in research-writing pedagogy because it accurately describes how successful introductions in practice tend to move, but no journal requires citing or explicitly following it.

Can the introduction state a hypothesis and an aim together?

Yes. Many papers state a general aim/objective and then an explicit testable hypothesis in the same closing paragraph, especially in experimental and quantitative work where a specific predicted relationship is expected. Check the target journal’s author guidelines for its specific convention.

How is a gap statement different from a limitation?

A gap statement, in the introduction, identifies what is missing in the existing literature before the study was conducted, and motivates why this study addresses it. A limitation, discussed in the Discussion section, identifies constraints on what this specific completed study can actually claim. They serve opposite ends of the paper: one motivates the work at the start, the other qualifies it at the end.

Should the introduction cite the same sources as the literature review?

Overlap is normal and expected — both draw from the same body of prior work. The difference is scope and purpose: the introduction cites selectively, only what’s needed to establish the territory and justify the gap, while a dedicated literature review (where the paper or thesis includes one as a separate section or chapter) aims for comprehensive synthesis of the relevant literature.