A large share of manuscripts are rejected before a single reviewer reads them. The editorial screen is fast, it is patterned, and almost every reason it fires is something you could have caught first.
Authors picture rejection as a verdict after peer review. A large fraction of manuscripts never reach a reviewer at all. In one journal's own accounting, nearly two in five submissions were desk-rejected without review ↗ — an editor read the abstract, skimmed the methods, and returned it in minutes. The reasons are predictable, and the clock they reset is not small: across biomedical journals, submission to publication runs from roughly 70 to 558 days ↗, and every rejection round starts it over.
A desk rejection falls into one of two buckets: the manuscript is a poor fit for this journal, or it is not yet ready for any. You cannot control the first from the writing; you can very nearly eliminate the second.
Out of scope. The commonest desk rejection is simply that the paper is not what this journal publishes — wrong specialty, wrong article type, wrong readership. An editor knows within a paragraph.
Insufficient novelty or importance for this venue. Sound work that does not clear the journal's bar for advance is redirected, not reviewed.
Self-check. Read the journal's aims-and-scope and its three most recent issues before you format a single table. Fit is a targeting decision, and it is cheaper to make before submission than to learn from a form letter. Choosing the right journal is not something an outside review can decide for you.
Incomplete methodological reporting. A design a reader cannot reconstruct — no eligibility criteria, no account of dropouts, no statement of how missing data were handled. Editors check this against the relevant reporting guideline, and a missing item is concrete and citable. Walk your design's checklist first: STROBE for observational studies, CONSORT for trials, TRIPOD for prediction models, PRISMA for systematic reviews.
A design that cannot support the claim. Causal language over an observational cohort, superiority implied by a single-arm series, equivalence asserted from a non-significant p-value. The claim is simply larger than the design can carry — and an editor sees it in the abstract.
Visible statistical problems. An underpowered study overclaiming a null, a conclusion that drifts from the results, numbers that do not reconcile between the abstract and a table. These are covered in depth in why manuscripts get rejected for statistics.
Administrative and ethics gaps. No ethics approval or registration statement, missing disclosures, figures below resolution, a manuscript over length, or writing an editor cannot follow. Individually minor; together, a signal the paper is not ready.
The readiness reasons above are all self-checkable. The reporting guideline is published; the design–claim question you can ask yourself; the statistical patterns are the ones a careful reviewer looks for; the administrative items are a list. Running them before submission converts a months-long round trip into an afternoon of edits.
A structured pre-submission review runs the methodological and statistical layer of that screen for you. RigorMD appraises a manuscript with two independent engines and a deterministic forensic layer, checks the design's reporting guideline item by item, and returns a severity-scored report grounded in your own quotes — across design–claim fit, results–conclusion alignment, numerical consistency, and reporting adherence. It flags the methodological reasons an editor would; it does not certify the manuscript, decide journal fit, replace peer review, or promise acceptance. See a full sample report →, read how the engine works, or review pricing — the pre-submission review is $25.