A desk rejection is a decision made without external review, usually by a handling editor, usually within days. It is the most common outcome for a submitted manuscript at selective journals, and the majority of desk rejections have nothing to do with the quality of the science.
That is worth restating, because it is the opposite of how most authors experience it. A desk rejection feels like a verdict on the research. It is usually a verdict on the match between the manuscript and the journal.
The eight causes, in the order editors encounter them
1. Article type mismatch
The journal does not publish the kind of thing you submitted, or publishes it only by invitation. Scoping reviews, methods papers, commentaries and registered reports are the frequent casualties. This check happens before anything else.
2. Scope mismatch
The topic sits outside what the journal covers, or inside a part of it the journal has quietly stopped publishing. The scope statement and the recent table of contents often disagree; the table of contents is the accurate one.
3. Missing declarations
Ethics approval, participant consent, funding, conflicts of interest, author contributions, data availability, and increasingly AI use. Several publishers now return manuscripts unreviewed when any of these are absent. This is a paperwork failure that reads as carelessness.
4. Format non-compliance
An abstract over the limit, too many keywords, the wrong reference style, single spacing where double is required, missing line numbers. Individually trivial. Collectively they signal that the paper was prepared for a different journal, which invites the question of who rejected it first.
5. Length
Over the word limit, or over the figure or table maximum. Some journals enforce this strictly at the desk; others allow an overage with justification. Check which yours is before assuming flexibility.
6. Language and readability
Not a matter of native fluency, but of whether an editor can follow the argument on a single reading. Papers that require re-reading to locate the claim are frequently returned, sometimes with a language-service recommendation attached.
7. No visible contribution
The paper reports competent work but never states what changes as a result. Editors are not obliged to infer your contribution. If it is not stated in the abstract and again at the end of the introduction, assume it will not be found.
8. Prior publication concerns
Substantial overlap with a preprint is usually fine and increasingly expected, but overlap with a published conference paper may not be. Similarity screening runs before review at most large publishers.
Which of these you can fix in an afternoon
Causes three through five are entirely mechanical. Cause one is a decision, not a fix. Cause two is a decision informed by twenty minutes of reading. Causes six and seven are writing problems that respond well to a targeted edit. Cause eight is a disclosure question you should resolve before submitting anywhere.
Six of the eight most common desk rejection causes can be resolved before you submit, without changing a single result.
Reading the rejection you did receive
Desk rejection letters are short and often templated, but the wording usually carries a signal.
- “Not a good fit for our readership” generally means scope or article type. Choose a different journal; revising will not help.
- “Does not meet our requirements for submission” generally means compliance. The same paper may be resubmittable once corrected, though check whether the journal permits that.
- “We receive many more submissions than we can publish” with no further detail is usually a priority decision rather than a fault. A comparable journal one tier down is often the right next step.
- Any specific comment at all is worth taking seriously. Editors who write a sentence of their own have spent time on your paper.
The practical takeaway
Run a compliance check against the specific journal before you submit, not after. Confirm article type first, scope second, declarations third. Those three checks take an hour and remove the majority of the risk that your paper is rejected by someone who never read your results.