The email came fast. Sometimes five days after submission, sometimes forty eight hours, and it fits in three short paragraphs: your manuscript will not be sent out for review. No reviewer reports, no line by line criticism, nothing you can actually work with. If you are reading this the same afternoon it arrived, take the rest of the day off. The decision is not going anywhere and neither is your paper.
Here is the part that matters once the sting fades: nobody evaluated your science. A desk rejection is one editor reading your title, your abstract and maybe the first lines of your method, and deciding that this manuscript would not survive the trip through their reviewers. That is a judgement about fit and first impression, made in minutes. Your analyses, your discussion and the argument you spent months building were never opened.
This guide is about the two weeks that follow. How to read a letter that seems to say nothing, what signals it does carry, how to run a one hour diagnosis on your own manuscript, how to choose the next journal without repeating the same mistake, and how much you genuinely need to rewrite before you press submit again.
There are two instincts after a desk rejection, and both cost more than the rejection itself. The first is to resubmit that same evening. You swap the reference style, redo the title page, send it to the next journal on your list, and eleven days later an almost identical email arrives. If the problem lived in the part the editor could actually see, sending that same visible part to a different editor buys you the same answer with a fortnight of waiting attached.
The second instinct is to reply and argue. Formally most journals accept appeals; in practice appeals against a desk rejection almost never work, because you are not disputing a factual error, you are disputing a judgement about fit and priority. There is one exception worth your time: if the editor clearly misread what your study is (treated an instrument validation as a pilot, or assumed an observational design where you ran a randomised trial), a short, unemotional message correcting that one point is legitimate. Anything longer reads as a complaint and gets a form reply.
It also helps to price what you actually lost. A desk rejection returns your manuscript in one to three weeks. A rejection after peer review can take four to six months and end in the same word, with reports attached. Measured in months of your life, the fast no is the cheap one. That is not a consolation phrase: it is exactly why experienced authors are willing to aim high, as long as the manuscript is clean enough not to fall for the avoidable reasons.
Desk rejection letters are written to be vague, and it is not personal. An editor at a busy journal signs dozens of them every week and cannot hand out diagnostics. Even so, the standard sentences are not random: they come from a small template set, and each one points somewhere. Here is what the usual phrases tend to mean, and what each one asks you to do next.
The most common one, and the best news in the set. It says your manuscript was read as belonging somewhere else, not as being weak. The fix is a journal decision, not a writing decision: submitted to a journal that has published two or three papers with your design and your population in the last two years, the very same file may go straight out to review. Do not rewrite the paper to answer this letter. Rewrite your shortlist.
Priority, novelty, limited advance, insufficient general interest: they all say the same thing. The editor believes the study is fine, and that it does not add enough for this particular journal. You have two clean routes. Either strengthen the contribution (an analysis that answers a question the field is currently arguing about, a comparison you had not modelled, an additional sample) or move one step down in impact, where a solid and correct study is precisely what the journal wants. Both are respectable. Inflating the contribution in a rewritten introduction is not, and editors read those paragraphs for a living.
The letter with no information at all. When this is everything you get, do not spend a week decoding the wording: play the base rates instead. The two dominant causes of desk rejection are scope mismatch and a weak first impression from title, abstract and visible method. Assume one of the two, run the diagnosis in the next section and act on whatever it turns up. That assumption is right more often than any reading you can squeeze out of a template sentence.
Take this one seriously: it is the only part of the letter written specifically about you. A transfer offer means the editor read enough to think the work belongs in that publisher’s ecosystem, just not at this level. Transfers usually carry your files across and sometimes any internal assessment, so the next decision comes faster. Accept if the target journal is properly indexed and its scope genuinely fits. Decline if it is an open access title you had never heard of and the article processing charge is the part of the email that gets the most space.
If the decision came back in under forty eight hours and mentions formatting, missing statements or file requirements, an editor was probably never involved. Editorial offices run a technical check first: word limits, required structure, ethics approval, informed consent, conflict of interest, data availability, reporting checklist (CONSORT, PRISMA, STROBE). This is the least painful rejection of all, because the repair is one afternoon of admin. It is also the most avoidable, which is why it is worth running that same check yourself before every submission from now on.
Before you decide anything about journals or rewriting, spend one hour testing your manuscript the way the editor did. Four checks, none of which require reopening your analyses.
Look at the method with the eyes of somebody hunting for a reason to say no, because that is the reading it got. Sample size justification, coherence between a cross sectional design and any causal verb in your conclusions, analyses that match the hypotheses you actually stated, assumptions genuinely checked rather than assumed. Across the manuscripts that have gone through our reviewer tool, most detected weaknesses (around 62% of them) sit in method and analysis. That is exactly the region an editor scans first, because it is where reviewers pile up their objections.
This is the decision that changes your odds the most. If your letter mentioned scope, it is the whole job. If it mentioned priority, it is half of it.
Fit is not thematic resemblance. A journal fits when it has recently published papers with your design, your kind of sample and your kind of question. Build a shortlist of three to five candidates and check each one properly before committing. Twenty minutes per journal saves you a month per journal.
If the letter pointed at priority or novelty, dropping one step usually solves the problem without touching a word of the manuscript: what is not exciting enough for a journal accepting 8% of submissions can be exactly what a solid second quartile journal is looking for. If the letter pointed at local interest, the issue is framing rather than level. A single country sample is not a defect; presenting it as though it only concerned that country is. The same study, framed against an international theoretical debate and discussed alongside samples from other contexts, reads as a contribution instead of a local report, and the data never changed.
This is the question that paralyses people. There are three honest levels, and your letter plus the one hour diagnosis tell you which one you are in.
For clean scope mismatches. The science stays exactly as it is. You rewrite the abstract so it stands alone, adjust the title so it states a finding, reframe the last paragraph of the introduction and the first of the discussion towards the new journal’s conversation, fix the formatting and write a new cover letter. Three to five working days, most of them spent on the abstract, which is fair because in the editorial filter the abstract is the paper.
For letters about priority, and for any diagnosis that showed the visible method looking thin. Here you add the sample size justification you never wrote, put effect sizes and confidence intervals in the abstract instead of bare p values, delete causal verbs your design cannot support, and make the introduction end on a gap your data genuinely fill. None of your results change. What changes is whether a sceptical reader in a hurry can see that you knew what you were doing.
The hardest case: the study is competently done and answers a question the field has stopped asking. Sometimes the repair is analytical (a moderation you had not modelled, a subgroup comparison that speaks to a live debate). Sometimes it is empirical (a second sample, a follow up). And sometimes the right call is to publish it in a good specialised or regional indexed journal, take the citation and move on. That is not defeat. A finished paper in a properly indexed journal beats a perfect paper you keep rewriting for two years.
Concretely, if the letter landed on Friday, here is a week that works. Monday: reread the letter cold, run the four tests, and write down in one sentence which of the three levels you are in. Tuesday and Wednesday: build the shortlist of three to five journals and pick one, using the checks above. Thursday and Friday: do the rewriting your level requires, starting with the abstract, always.
The following Monday, make the last pass as a stranger. Read only the title, then the abstract, then the method, in that order, and at each step ask whether you would send this out for review. Then write the cover letter: two paragraphs answering the only question the editor really has, which is why this paper, in this journal, now. Connect your contribution to one or two recent papers the journal itself published. Generic templates get skimmed; a letter that proves you read the journal does not.
And then send it. The most expensive habit after a desk rejection is not resubmitting too fast, it is not resubmitting at all: manuscripts that sit in a folder for eight months while their authors wait to feel ready. Your paper was rejected in five days by somebody who never read your results. That is information about fit and about first impression, and you now know what to do with both.
No. There is no shared registry between publishers: each journal only sees its own history. A desk rejection at one journal has no effect whatsoever on your chances at another, and you are under no obligation to mention it. Within the same journal there is an internal record, which only matters if you resubmit the same manuscript with cosmetic changes, something editorial systems detect easily.
Long enough to do the work your diagnosis calls for, and no longer. For a clean scope mismatch, three to five days is honest. For a rebuilt front section, two to three weeks. Waiting months does not improve the manuscript; it just moves your data further away from the literature it is arguing with.
Rarely. Appeals succeed when there is an objective misunderstanding you can demonstrate in one paragraph, for example that the editor treated your randomised trial as an observational study. If your appeal amounts to explaining why the work matters more than the editor thought, the week it costs you is worth far more invested in the next submission.
Only if the letter invites it (some say they would consider a substantially revised version) or if the manuscript genuinely changed: new framing, additional analyses, a different central claim. Resending the same file with a new title page usually produces a second rejection, faster than the first.
It can be a factor, but not in the way most authors fear. Editors do not reject you for an accent in your prose or for imperfect articles. They reject when they have to reread a sentence to work out what you did. The test is the abstract: if a reader has to reconstruct your design from context, that is worth fixing, and a professional language edit before submission costs a fraction of three months lost between two journals.
You are not. Desk rejection is the most frequent editorial decision at journals with high submission volume. Rates are commonly reported above 50% of submissions at large generalist high impact journals, and somewhere between 20% and 40% at specialised psychology and health titles. Most of the papers you admire were rejected somewhere first, often without review.