You submitted the manuscript, you braced yourself for three months of peer review, and six days later a four sentence email arrived: the journal will not send your paper to reviewers. That is a desk rejection, and in high impact journals it is the most common editorial decision there is. No reviewer read your work. One person, an editor with a queue of forty other submissions, read your title, your abstract and the first lines of your method, and stopped.
The useful thing to understand is what that decision actually measures. It is not a verdict on your science, because your science was never read. It is a verdict on fit and on first impression: does this belong in this journal, and does the visible surface of the manuscript survive five skeptical minutes. Those are repairable in a week. Your study design is not.
This guide covers what a desk rejection is, how fast it arrives, how often it happens, the seven reasons editors actually give when they explain their criteria, how it differs from a rejection after peer review, and what to do with the manuscript before you send it anywhere else.
A desk rejection (also called an editorial rejection, or a rejection without review) is the decision by an editor not to send your manuscript out for peer review at all. The editor in chief, or a handling editor in your subject area, reads the submission and closes it. No external referee is involved, no reports are produced, and nothing about your analysis is evaluated.
The clearest signal is speed. Desk rejections typically arrive between 48 hours and three weeks after submission. If a decision reaches you in the first week, it almost certainly never left the editor’s desk. Decisions that take three months or more usually mean reviewers were involved, even when the letter arrives with no reports attached. A second signal is the absence of any specific criticism: a desk rejection letter tells you what will not happen next, not what is wrong with page 12.
It helps to know how little time you are competing for. Editors at busy journals screen submissions in batches, often giving each one somewhere between two and ten minutes on the first pass. In that window they read the title, the abstract, the cover letter if there is one, and they skim the method for sample and design. Everything you spent eighteen months on is being judged by the two percent of it that is visible in that window.
Exact rates are rarely published journal by journal, so be skeptical of any source that gives you a single confident figure. What editors and publishers do report, in editorials and author guidance, falls into a fairly consistent range. At large general journals with very high submission volumes, more than half of all manuscripts are commonly reported to be rejected without review, and at the most selective titles that share rises well above 60 percent. At specialized journals in psychology and health sciences, the range usually reported sits somewhere between 20 and 40 percent of submissions.
The pattern behind those numbers is easier to trust than the numbers themselves: the lower a journal’s overall acceptance rate, the larger the share of its rejections that happen before peer review. Reviewers are unpaid, slow to recruit and easy to burn out. An editor who sends out a manuscript that two referees will reject in six weeks has spent a scarce resource for nothing, so the screening filter gets stricter as the queue gets longer.
For you, this reframes the odds in a way that is actually encouraging. If a journal desk rejects 40 percent of submissions and then accepts about a third of what goes to review, the first gate is the one you can influence most cheaply. Passing it does not require better data. It requires a manuscript whose scope, abstract and method read as a credible fit in the first five minutes.
Editors almost never state the real reason in the letter, but when they explain their screening criteria in public (journal editorials, author webinars, conference panels) the same seven reasons come up. They are listed here from most to least frequent.
This is the single biggest cause, and it is not close. The study can be excellent and still answer a question the journal does not publish: an applied clinical trial sent to a basic cognition journal, a questionnaire validation sent to a title that publishes only interventions and reviews. The aims and scope page exists precisely to prevent this, and it is the page authors most often skip. Read it, and read the last four issues, before you decide.
Going 40 percent over the word limit, using a different reference style, skipping the required structured abstract, or leaving out mandatory statements (ethics approval, informed consent, conflicts of interest, data availability, reporting checklists like CONSORT or PRISMA) sends a message editors read instantly: this author did not read the instructions, so this author will probably be difficult during revision. It is the most avoidable reason on the list and it still costs papers every week.
The bar is not native fluency or perfect idiom. The bar is whether the editor has to reread a sentence to work out what you did. When an abstract demands that effort, the editor assumes reviewers will not make it, and declines. If you write from a non anglophone lab, professional language editing before submission costs a small fraction of what three months lost between two journals costs you.
Experienced editors diagnose method problems without opening the method section: 60 participants for a structural equation model, a cross sectional design with causal claims in the title, an abstract that reports p values and no effect sizes, a convenience sample presented as representative. They do not need to read further, because they already know what the reviewers would say.
Our own data points the same way. Across the manuscripts run through the paper reviewer tool, roughly six in ten of the weaknesses flagged sit in method and analysis. That is exactly the region an editor scans first, because it is the region where objections cannot be written away.
The work is competent but confirms what the field already accepts, or repeats a design published many times without acknowledging it. Q1 journals do not publish correct studies, they publish studies that move something. If your introduction cannot honestly complete the sentence “this is the first study to…” or “unlike previous work, here we…” with real content, the editor will notice before you do.
A single country sample is not a problem. Presenting it as if it only matters in that country is. “Burnout among nurses at a hospital in Ankara” reads as a regional paper. The same dataset, framed against a theoretical debate that international readers are having, and discussed against samples from other contexts, reads as a contribution. The data does not change. The introduction and discussion do.
Nearly every indexed journal runs submissions through iThenticate or an equivalent before an editor sees them. High text overlap, including with your own earlier papers (self plagiarism), undeclared preprint or duplicate submission, image irregularities, or signs of salami slicing produce an immediate desk rejection and sometimes a note in the publisher’s records. Paraphrasing your own published method section is not enough. Rewrite it.
Reason 4 deserves more detail, because it is the one authors most often mistake for bad luck. An editor who reads past the abstract skims the method and the first table looking for what is missing rather than for what is there. Four patterns end the read early, and none of them needs a reviewer to be noticed.
The stated aim and the chosen design belong to different categories: a causal aim (“to determine the effect of X on Y”) with a cross-sectional correlational design, a predictive aim with a small sample and fifteen predictors, a psychometric aim (“to validate the structure of the questionnaire”) without a single confirmatory factor analysis. This is not the same as executing a correct design poorly, which is fixable. A mismatch between aim and design is structural: you cannot report your way out of it, you have to change the analysis or scale back the conclusions to what the design supports.
The details that are most often missing: an a priori sample size justification (a power analysis or a simulation), an explicit treatment of missing data (full information maximum likelihood, multiple imputation, or why listwise deletion is defensible here), instruments validated in the population you studied, and ethics approval with its reference number. “We collected as much data as we could” is the weakest sentence a method section can contain. If your final n fell short, say so and add a sensitivity analysis showing the smallest effect you could reliably detect: that turns a weakness into a managed limitation, which is what an editor wants to see.
Unchecked assumptions (normality of residuals, homoscedasticity, sphericity, multicollinearity), p values without effect sizes or confidence intervals, uncorrected multiple comparisons, and numbers in the text that do not match the tables. Each of these alone is a minor comment. Several together tell the editor that the team does not fully command its own analysis, and that weighs more than any single error. Formatting matters for the same reason: exact p values to three decimals except p < .001, tables with a note and consistent decimals, figures with a descriptive title. It is the cheapest signal an editor has of the care behind the numbers.
Causal language (“our results show that X causes Y”) when the design only supports association, generalization to populations the sample does not represent, and hypotheses presented as confirmed when the analyses support them more weakly than the text suggests. The discussion is where it becomes clear whether the team understands the limits of its own design. Match the language to the design, not to what you hoped to find.
These two decisions arrive in similar looking emails and mean almost opposite things, so they call for opposite responses. A rejection after peer review comes with reports from two or three referees who read the whole manuscript. It hurts more and it is worth far more: those reports tell you precisely what to strengthen, and a large share of papers published in good journals were rejected somewhere else first, with reviewer comments that improved them.
A desk rejection gives you almost none of that. What it does tell you, fairly reliably, is that you failed on fit or on first impression: wrong journal, or an abstract and method that did not survive a five minute read. The depth of your work remains unevaluated, which is why the correct response is rarely to redo the analysis and usually to reconsider the target and rewrite the visible surface.
There is also a difference in what each one costs. A desk rejection returns your manuscript in one to three weeks. A rejection after review can take four to six months. That asymmetry is why some experienced authors deliberately aim one tier above where they expect to land: the time cost of trying is small, provided the manuscript is clean enough that it will not fall for reasons 2, 3 and 7, which are entirely within your control.
The standard letter is vague by design: “after careful consideration, we have concluded that your manuscript is not suitable for our journal”, or “given the volume of submissions, we can only send out papers of the highest priority”. Editors write dozens of these a week and will not risk an argument by being specific. Even so, the wording leaks information.
If the letter mentions suitability or scope (“not within the scope of this journal”, “better suited to a more specialized outlet”), you are looking at reason 1, and the fix is journal selection, not rewriting. If it mentions priority or advance (“does not meet our priority threshold”, “limited advance over existing literature”), that is reason 5, and you need to strengthen the contribution before you resubmit anywhere. If the letter offers a transfer to another journal in the same publishing group, take it seriously: that is an editor telling you where your work does fit, and transfers often carry your files and sometimes speed up review.
And if the letter says nothing interpretable at all, assume the base rate. Scope mismatch and weak first impression account for most desk rejections, so that assumption is right more often than any other guess, and it produces the most useful plan.
The most common mistake is to resubmit the same file to another journal the next morning with the references reformatted. If the problem was pure scope mismatch, that can work. If the problem was first impression, you have just bought a second desk rejection and another three weeks of waiting. Give it five working days and do these things in order.
If the rejection cited priority or novelty, moving down one impact tier often solves it without touching the manuscript: what is not a priority for a journal accepting 8 percent of submissions can be exactly what a solid second quartile title is looking for. If it cited local interest, reframing matters more than tier. Keep the sample, change the framing so an international reader sees why your context tests something general.
Do not let the pipeline stall while you decide. Across the manuscripts reviewed with our tool, more than eight in ten received the equivalent of a major revisions verdict, which is a reminder that surviving the editor is only the first gate. The objections waiting on the other side are the same ones you can read now, for free, while you still have time to answer them in the manuscript instead of in a rebuttal letter.
The surest way to anticipate a desk rejection is to have someone read the full manuscript with a reviewer’s eye before the editor does, and tell you precisely what would fail. That is what Previum (AI peer review) is for. Upload the manuscript, optionally with your aims and the target journal, and it reads it the way a demanding reviewer would: scope and contribution, the fit between aims and design, the visible method, the reporting.
The Basic review shows the two findings that decide the review, each with the quote from your text and how to fix it, plus checks anchored to your text: recalculated p values, figures that contradict each other, and references verified one by one. The Professional review adds two complete reviewers, a simulated editor’s letter and journals that fit. It does not replace the journal’s peer review, which is done by people with their own judgment; it lets you arrive at it with fewer surprises, and it costs you minutes instead of the three weeks a desk rejection takes to arrive.
Usually between 48 hours and three weeks. A decision in the first week is almost always a desk rejection. If the decision takes three months or more, reviewers were probably involved even if no reports are attached. Some journals state a target screening time in their author guidelines, which is worth checking before you start refreshing your inbox.
It depends heavily on the journal. Large general journals with very high submission volumes commonly report rejecting more than half of submissions without review, and the most selective titles report considerably more. Specialized psychology and health journals more often fall in the 20 to 40 percent range. Treat any single precise number you see online with suspicion: most journals do not publish this figure.
There is no shared blacklist across publishers. Each journal sees only its own history, so a desk rejection at one title has no effect on your chances at another. Within the same journal there is an internal record, which matters mainly if you resubmit the same manuscript unchanged, since editorial systems flag that automatically.
Formally, almost every journal allows it. In practice appeals against desk rejections rarely succeed, because the decision is about fit and priority rather than a factual error you can disprove. It is worth trying only if you can point to an objective misunderstanding, for example the editor classifying your study design incorrectly. Otherwise the week you would spend appealing is worth more spent on the next submission.
Only if the letter explicitly invites it (some say they would consider a substantially revised version) or if the manuscript has genuinely changed: new framing, additional analyses, a different research question. Resubmitting the same file with cosmetic edits usually produces a second rejection faster than the first one.
No. It means the paper did not pass a fit and first impression filter that lasted minutes. Nobody assessed your analyses or your discussion. Manuscripts rejected without review at one journal are routinely published, with no changes to the science, at a better matched one. The signal is not that the study is weak, it is that the journal choice or the visible surface of the paper needs work.
Check for recognizable indexing (Scopus, Web of Science, or DOAJ for serious newer open access titles), an editorial board of identifiable academics, and a consistent publishing history. Be wary of near-automatic acceptance combined with high fees. Think. Check. Submit. and your university library are good starting points. A desk rejection costs you weeks; a predatory journal costs you the paper.
No, and it does not try to. It is a check before you submit: it shows you the findings that would decide the review, with the quote from your text and how to fix them, so you reach the real editor with fewer surprises. The journal’s review is still done by people, with their own judgment, and only they decide.
It is less informative but much cheaper. You lose two or three weeks instead of five months, and you keep the option of submitting elsewhere almost immediately. The real cost is that you learn nothing about the quality of the work, so you have to generate that feedback yourself before the next attempt.