Run this before you submit, not after the rejection.

What this checklist is for

A desk rejection is a rejection an editor issues before the manuscript reaches a reviewer. It happens at the editor's desk, on a first read, and it is decided by a small set of signals that show up in minutes rather than by a close reading of your method.

The reasons a paper gets desk rejected are covered well elsewhere. This page is the operational companion. It is a list to run against a specific manuscript, in the days before you submit, organized by what you can still change and how much time each check costs you. Where an explanation of desk rejection tells you why a reason exists, this page tells you what to do about it this afternoon.

Work through the sections in order. Fast checks first, because they catch the most common problems for the least effort. Slower checks next. Then a short section on what a checklist cannot fix, because pretending it can is its own kind of failure.

The full list takes most authors somewhere between an afternoon and two days, depending on how many collaborators need to weigh in on the day-long items. That is short compared to a revise-and-resubmit cycle, which is usually measured in months. Running the checklist once, seriously, before submission is close to the cheapest change you can make to how quickly a paper moves.

The five-minute desk rejection scan

Start here. None of these require rereading the paper. They require reading three or four specific things you already have open.

• •Open the journal’s current aims and scope statement and read it once, straight through. Does your paper match the sentence, not the general field it sits in?
• •Open the table of contents of the journal’s last two issues. Does anything resemble your paper in design or scope?
• •Reread your own abstract as if you had never seen the manuscript. Can you find the research question in the first two sentences?
• •Check the word count against the journal’s stated limit, not the limit of a journal you are more familiar with.
• •Confirm the ethics approval statement is present in the methods section, with an approval number or an explicit exemption.

Scope mismatch and missing ethics statements account for a large share of desk rejections, and both are checkable in minutes because neither requires new analysis. If either one fails here, fix it before you spend time on anything below.

The afternoon checks

These take longer because they ask you to verify something against the manuscript itself, not just read a policy page. Budget an afternoon, not a week.

• •Trace every number that appears in your abstract back to the table or figure it came from. A mismatch here is one of the fastest ways to lose an editor’s confidence.
• •Check that your stated research question and your actual analysis are the same question. A study that drifted during data collection sometimes ends up answering something narrower than the introduction promises.
• •List every statistical test used in the results and confirm each one matches the data type and design described in the methods. A test that assumes independence run on repeated measures from the same subject is the kind of thing an editor catches on page four.
• •Search your reference list for papers published in the target journal in the last two years. Their absence reads as a sign you have not checked whether the journal already covers this ground.
• •Read the discussion section and mark every sentence that claims more than the results support. Cut or soften each one.

The reference-list check is worth doing even when it feels unnecessary. An editor who edits the journal you are submitting to notices, within seconds, whether you have engaged with work the journal itself has published.

The checks that take a day

A smaller set of items needs a full day because they involve outside input, not just your own reread.

• •Ask a colleague outside your immediate collaborators to read only the abstract and tell you, in one sentence, what question the paper answers. If their sentence does not match your intended question, the abstract is the problem, not the reader.
• •Run your study design past a checklist appropriate to the design: CONSORT for trials, STROBE for observational studies, PRISMA for systematic reviews. These exist to catch exactly the flaws editors are trained to spot on a first read.
• •If a statistician has not looked at the analysis since data collection ended, send them the methods and results sections and ask specifically whether the tests match the design. This is a narrower ask than a full review and gets a faster answer.
• •Reread the cover letter separately from the manuscript. It should state the contribution in plain language and should not restate the abstract at length.

What increases desk rejection risk that you can't fix this week

Some findings from this checklist point at problems a checklist cannot solve. Naming them plainly is more useful than pretending an afternoon of edits will handle them.

A study design that cannot answer the question you are asking is one of these. If the check above on tests-matching-design turned up a cross-sectional dataset being used to support a causal claim, no amount of rewording the discussion fixes that. The honest options are to narrow the claim to what the design supports, submit to a venue with a different standard for that kind of evidence, or collect more data. None of those happen this week.

A genuine scope mismatch, once you have actually read the aims and scope statement rather than skimmed it, is the same kind of problem. The fix is choosing a different journal, not editing the introduction until the paper sounds like it belongs somewhere it does not.

A worked example

Consider a manuscript on a six-month exercise intervention in adults with prediabetes, submitted to a general endocrinology journal. Running the checklist above surfaces three things.

First, the abstract leads with three sentences on the global burden of diabetes before stating, in the final sentence, that the study measured change in insulin sensitivity after a structured walking program. The five-minute scan catches this: a colleague asked to summarize the question in one sentence describes the burden of diabetes, not the study. Fix: move the research question to the second sentence and cut the burden framing to one clause.

Second, the results report a within-group paired comparison for the intervention arm and a separate between-group comparison for the control arm, but the methods section only pre-registers the between-group test. The afternoon check on matching tests to design catches this. Fix: either report the within-group result as exploratory and clearly labeled as such, or remove it if it was not planned.

Third, the reference list has nothing published after year one of what became a three-year data collection window. The afternoon check on recent literature in the target journal catches this. Fix: a focused search covering the two most recent years, added to the introduction and discussion where relevant.

None of the three required new data. All three were visible from the manuscript itself, once someone looked with the specific question a checklist asks rather than a general reread.

What this checklist does not do

A checklist finds what you already know to look for. It does not surface problems in a research question you are too close to see clearly, and it does not replace a reader who understands your specific subfield well enough to know that a competing explanation exists that you have not addressed. Those require a second set of eyes, whether a colleague, a statistical consultant, or specialist reviewers who read the whole manuscript.

Treat this page as the floor, not the ceiling. Clearing every item here removes the most common reasons an editor rejects a paper without sending it out. It does not guarantee acceptance, and it is not meant to.

There is also a limit to how much any list can adapt to a specific field. A checklist built for general use has to describe checks broadly enough to apply across disciplines, which means it will miss field-specific conventions your target journal cares about: a preferred reporting format in one subfield, an expected sample-size justification in another. Add those to the list yourself once you know your journal well enough to name them.