An outline is the single highest-leverage step in preparing a research manuscript, and it happens before a single full paragraph of the paper gets written. Researchers who draft section-by-section from a skeleton — rather than writing linearly from the first sentence — spend less time in unproductive rewriting later, because the argument’s logical structure gets tested while it is still cheap to change. This guide covers how to build that skeleton: the standard IMRaD structure used across most empirical fields, a section-by-section method for populating it, how the outlining task itself changes across an empirical STEM paper, a humanities argumentative paper, and a systematic review, and two concrete techniques — reverse-outlining and using headings as a working skeleton — for applying all of this in practice.
Why Outline Before Drafting a Manuscript
A manuscript outline is not a formality for its own sake. It does three specific things a linear first draft cannot:
- It separates argument structure from prose. Deciding what belongs in the Discussion versus the Results, or whether a finding needs its own subsection, is a structural decision. Making it inside full paragraphs means every structural change requires rewriting sentences, not just moving bullet points.
- It matches the paper to what a journal — and its reviewers — actually expect to find. The International Committee of Medical Journal Editors (ICMJE) Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals state plainly that “the text of articles reporting original research is usually divided into Introduction, Methods, Results, and Discussion sections” — a structure widely adopted well beyond medicine. A reviewer scanning a submission for this shape and not finding it is a friction point an outline exists to prevent.
- It surfaces gaps before they cost a revision cycle. An outline that lists every subsection and, ideally, every intended table and figure exposes missing methodological detail, an unsupported claim, or a results section with no corresponding discussion point — while it is still a list, not a manuscript.
See CASRAI’s Scholarly Writing pillar for the broader manuscript-preparation and publication landscape this guide sits within.
The Standard IMRaD Structure
IMRaD — Introduction, Methods, Results, and Discussion — is the default organizing structure for original empirical research articles, and the one a research-paper outline is built around by default unless a target journal or discipline specifies otherwise. Historical research by Sollaci and Pereira (The Introduction, Methods, Results, and Discussion (IMRAD) Structure: A Fifty-Year Survey, Journal of the Medical Library Association, 2004) traces its adoption: the structure was already proposed as an ideal for scientific writing in the early twentieth century, but authors did not consistently follow it until international scientific-publishing conferences after World War II recommended it; by the 1970s–80s it had become the dominant format in leading medical and physics journals, and it remains the default reference point behind the ICMJE Recommendations today.
A complete manuscript outline built on IMRaD typically has more scaffolding than the four-letter acronym implies. In order, a full outline usually includes:
- Title — precise and specific enough to be found and cited correctly; some journals additionally require a short running title (often capped around 40 characters).
- Abstract — per ICMJE, original research, systematic reviews, and meta-analyses require a structured abstract covering background/context, purpose, basic procedures, main findings (with effect sizes and significance where possible), and principal conclusions.
- Introduction
- Methods (sometimes “Materials and Methods”)
- Results
- Discussion (often followed by a short, separate Conclusion)
- References, tables, figures, and supplementary material
Outlining, in practice, means drafting a one- or two-sentence statement of purpose for each of these sections before writing full prose for any of them — plus a placeholder list of every table and figure the Results section will need, since planning those early avoids restructuring the Results text around figures added later.
Section-by-Section Outlining Approach
Each IMRaD section has a distinct job. Outlining a research paper well means writing down, in a sentence or two per subsection, exactly what that section’s job requires — before drafting the full text.
Introduction
Per the ICMJE Recommendations, the Introduction should “provide a context or background for the study” and “state the specific purpose or research objective of, or hypothesis tested by, the study or observation.” A working outline for this section typically moves through three narrowing steps: (1) the broad problem area and why it matters, (2) the specific gap in existing knowledge the paper addresses — often built on a short literature synthesis, see below — and (3) the paper’s specific objective, question, or hypothesis, stated explicitly enough that a reader can predict what the Methods section will do.
Literature Review
In IMRaD-structured papers, the literature review is usually not a separate labeled section but a compressed synthesis embedded in the Introduction — enough to establish the gap the paper fills, not an exhaustive survey. (A standalone, comprehensive literature-review section or article is a different genre with its own structure; see CASRAI’s comparison of literature review vs. systematic review vs. scoping review for how these diverge.) When outlining this piece, list the 3–6 specific prior findings or positions the paper is responding to, and for each one, note in one clause what it does or doesn’t establish — that list becomes the connective tissue between the broad problem statement and the paper’s specific objective.
Methods
ICMJE’s guiding principle for this section is that it “should aim to be sufficiently detailed such that others with access to the data would be able to reproduce the results.” An outline for Methods should enumerate, as subheadings or bullet points, each discrete methodological component the finished section will need to cover: study design, participant or sample selection criteria, data collection procedures/instruments, and the statistical or analytical approach — in whatever order the discipline’s conventions expect. Outlining this section carefully is disproportionately valuable, since a Methods section reconstructed after the fact from memory is a common source of reviewer requests for clarification.
Results
ICMJE instructs authors to “present your results in logical sequence in the text, tables, and figures, giving the main or most important findings first,” and specifically to avoid repeating in prose everything already shown in a table or figure. A Results outline is best built as a numbered list of findings in the order they will be presented, each one tagged with which table, figure, or statistic supports it — this is the point at which deciding “does this belong in a table or in the sentence” is cheapest to get right.
Discussion
ICMJE describes the Discussion as the section that should “briefly summarize the main findings, and explore possible mechanisms or explanations for these findings,” while also requiring authors to state the study’s limitations and its implications for future research and practice. A useful outlining discipline here is to map each Discussion point back to a specific Results-section finding one-to-one — a Discussion outline with claims that don’t trace back to anything in Results, or Results findings with no corresponding Discussion point, is exactly the kind of structural gap an outline is meant to catch before drafting.
How Outlining Differs by Discipline and Paper Type
IMRaD is the default, not a universal template. The outlining approach itself needs to change depending on what kind of paper is being written.
Empirical STEM Paper (IMRaD)
For an empirical study in the biomedical, life, physical, or social sciences reporting original data, the IMRaD outline described above applies close to as-is. The outline is built around demonstrating that a specific, falsifiable question was tested with a reproducible method and answered with evidence — the four sections exist specifically to let a reader and reviewer evaluate that chain efficiently. The ICMJE Recommendations note that non-original-research article types — meta-analyses, case reports, reviews, editorials — may need a different organizational format even within the same broadly empirical fields.
Humanities Argumentative Paper
A humanities paper — literary analysis, historical argument, philosophical essay — is not organized around Methods and Results in the empirical sense, because it is not reporting a reproducible procedure applied to a dataset. Instead, the paper is organized around building and defending a thesis. An outline for this genre is typically built around: (1) the thesis statement itself, stated as precisely as an empirical hypothesis would be, (2) each major line of supporting evidence or textual/archival analysis as its own section, ordered by the logic of the argument rather than chronology of research activity, and (3) explicit engagement with the strongest counter-argument or competing interpretation, since humanities argumentation is expected to anticipate and address objections rather than simply accumulate supporting points. There is no single cross-disciplinary structural standard equivalent to IMRaD here — expect the outline’s shape to be governed by the specific sub-field’s conventions and the target journal’s house style rather than one universal template.
Systematic Review
A systematic review is closer in outline shape to an empirical IMRaD paper than to a narrative literature review, but its Methods and Results sections describe a review process rather than a primary study. Systematic reviews and meta-analyses are typically structured and reported against the PRISMA 2020 statement (Page MJ, et al., “The PRISMA 2020 statement: an updated guideline for reporting systematic reviews,” BMJ 2021;372:n71), which specifies a 27-item checklist and a flow diagram tracking records identified, screened, and included. Outlining a systematic review means working from that checklist directly: an eligibility-criteria subsection, a search-strategy subsection (databases, date ranges, search terms), a study-selection and data-extraction subsection, a risk-of-bias assessment subsection, and — if the review includes pooled quantitative analysis — a separate synthesis/meta-analysis subsection, distinct from a narrative summary of individual studies. See CASRAI’s dedicated guide to PRISMA and systematic review methodology and the systematic review vs. meta-analysis comparison for the full checklist and how the two relate.
Practical Outlining Techniques
Reverse-Outlining a Draft
Reverse outlining is not a pre-drafting technique but a revision technique — it is included here because it is the most effective way to check whether an outline actually held up once a full draft exists. The method, as taught by university writing centers, is straightforward: go through a completed draft paragraph by paragraph (or subsection by subsection, for a full manuscript) and, without looking back at prose detail, write down one sentence summarizing what each paragraph or subsection actually is about and, separately, what it does in the paper’s argument. The resulting list is functionally the outline the draft actually produced — as opposed to the outline the writer intended. Comparing the two surfaces exactly the kind of structural drift that is invisible while reading prose linearly: a paragraph doing two jobs at once, a subsection that doesn’t map to anything on the original outline, or a Discussion point (see above) with no corresponding Results finding. Reverse-outlining a full manuscript draft before submission is a fast, low-cost check against exactly the structural gaps a reviewer is most likely to flag.
Using Headings as a Skeleton
The most direct way to turn an outline into a working document is to write the full heading hierarchy first — Introduction, Methods (with its planned subheadings already in place), Results, Discussion — and only then fill in prose underneath each one. This keeps the structural decisions (what goes where, what the section order communicates) separate from the sentence-level writing, and it means a manuscript is never in a state where reorganizing requires rewriting rather than simply moving a heading and its content block. It also makes gaps immediately visible: an empty heading with nothing under it is a much clearer signal than a missing idea buried somewhere inside a long, undifferentiated paragraph. For most word processors and reference managers, working from a heading skeleton also makes it straightforward to generate a working table of contents to sanity-check overall balance — a bloated Introduction next to a thin Discussion is easy to spot as a heading list, much harder to spot by reading start to finish.
Frequently Asked Questions
Does every research paper need to follow IMRaD?
No. IMRaD is the default for original empirical research articles and is what most journal templates and reviewer expectations are built around, but the ICMJE Recommendations themselves note that other article types — reviews, meta-analyses, case reports, editorials — commonly use a different structure, and humanities/argumentative papers are organized around a thesis rather than a method-and-results structure at all. Check the target journal’s author guidelines before assuming IMRaD applies.
How long should a research paper outline be before drafting starts?
There’s no fixed standard length. A useful working outline is detailed enough that every planned subsection, table, and figure is listed with a one-sentence description of its content — typically well short of a page for a standard journal article, but long enough that nothing in the eventual draft should come as a structural surprise.
Where does the literature review go if a paper doesn’t have a separate literature-review section?
In a standard IMRaD paper, the literature synthesis is compressed into the Introduction, providing just enough context to establish the gap the paper addresses. A full standalone literature review — narrative, systematic, or scoping — is a distinct paper genre with its own structure; see CASRAI’s comparison of the three review types linked above.
Should figures and tables be planned during outlining or added later?
Planning them during outlining is strongly preferable. Deciding early which findings belong in a table or figure versus in-text prose (per ICMJE’s guidance to avoid repeating tabulated data in the text) shapes how the Results section itself should be written, not just how it should be illustrated afterward.
Is reverse-outlining only useful for revision, or can it help while outlining a new paper?
It’s primarily a revision check, but the same one-sentence-per-section discipline is worth applying to a forward outline too: after drafting an initial outline, try summarizing what each planned section is supposed to do in one sentence. If that’s hard to do, the section’s purpose likely isn’t well defined yet.
Related CASRAI Resources
- Scholarly Writing pillar — the full manuscript-preparation and publishing-craft cluster this guide belongs to
- PRISMA and Systematic Review Methodology
- Literature Review vs. Systematic Review vs. Scoping Review
- Systematic Review vs. Meta-Analysis
- Systematic review (dictionary term)
- Peer Review: What It Is, How It Works, and Why It Matters
- What Is a Dissertation? Structure, Defense, and Deposit
- COPE Guidelines Explained
