Clicking submit takes a second. Finding out it was premature takes two weeks if the editor rejects it from the desk, and four to six months if it goes to review first. Either way, the problem you could have fixed in an afternoon has now cost you a season of your academic year, and in most cases it was not your data that failed. It was something visible on page one.
Most pre-submission checklists are ordered the way a manuscript is written: title, abstract, introduction, method, results, discussion, references. Editors do not read that way. They read for reasons to stop, and they find those reasons in a predictable order. This checklist follows the editor's order instead: journal fit first, because it kills more submissions than everything else combined, and the fine detail of your discussion last, because no editor rejected a paper at the desk over paragraph 43.
Work down it in sequence. If you cannot tick item one, nothing below it matters yet: you are polishing a manuscript for a journal that was never going to publish it. If you can tick everything down to the last section, you are not submitting early, you are submitting ready.
Your paper faces two very different gatekeepers, and almost every author prepares for the wrong one first. The second gatekeeper is the reviewer, who spends two or three hours with your full manuscript. The first is the editor, who spends five to fifteen minutes and decides whether reviewers ever see it at all. That first decision is the desk rejection, and it is the most common editorial outcome at selective journals: at high-volume generalist titles more than half of submissions never reach review, and at specialised psychology and health journals the figures commonly reported sit somewhere between 20% and 40%.
The editor is not being lazy. Peer reviewers are unpaid, scarce and increasingly hard to recruit, so an editor only spends them on manuscripts with a real chance of surviving. That means the desk decision is made on a small, fixed set of signals: does this belong here, did the author follow the rules, does the abstract show a contribution, does the method look defensible, can I read the English without effort. Six or seven signals, all of them visible before page three.
The good news for you is that every one of those signals is fixable in days. The bad news is that they are exactly the signals authors leave until last. In the 346 manuscripts we have analysed with our reviewer tool, 62% of the weaknesses detected sit in method and analysis, and 84% of the manuscripts would have received a Major revisions verdict. Those are not exotic problems. They are the same problems an editor spots in five minutes and a reviewer spends five pages describing.
When editors write about why they reject at the desk, in editorials, author guidelines or conference panels, scope is almost always the first reason they name. A methodologically clean paper sent to the wrong journal loses to a weaker paper sent to the right one, because the editor is not ranking science in the abstract, they are asking whether their readers signed up for this. A questionnaire validation sent to a journal that publishes only trials and reviews is not a borderline call for them, it is a two-minute decision.
Fit is not thematic resemblance. It means the journal has recently published papers with your kind of question, your kind of design and your kind of sample. Before you commit, build a shortlist of three to five candidates and run each of them through these five checks.
Here is the fastest version of the whole check. Write two sentences: this journal published X and Y in the past two years, and my paper extends that specific conversation because Z. If you can write them with real titles and a real Z, you have fit, and those two sentences are also the strongest paragraph of your cover letter. If you cannot, you do not have a writing problem, you have a target problem, and no amount of editing will fix it.
Nobody ever got published for formatting correctly, but manuscripts sink every week for ignoring the instructions. This is the most frustrating category of rejection because avoiding it requires no talent, no new analysis and no coauthor meeting: it requires one boring afternoon. It also carries a signal the editor reads instantly, which is that an author who did not read the guidelines will probably not be careful when answering reviewers either.
Reread the author guidelines on the day you submit, not the day you started writing. They change without announcement, and following a version from two years ago produces administrative rejections that never even reach a human judgement.
The abstract is the most read and least revised part of any manuscript. It gets written last, usually at midnight, and rarely gets a second pass. That is a bad allocation of effort twice over: the editor decides with it now, and after publication it is the only thing most people who find your paper will ever read.
Write it as if it were the whole submission, because at the desk it is. An abstract that forces the editor to open the PDF to find out what you actually found is an abstract that failed.
Experienced editors diagnose methodological problems without opening the method section, because the abstract leaks them: 60 participants for a structural equation model, six predictors on a sample of 47, a cross-sectional design with causal language in the title, p values with no effect sizes anywhere. They do not need to read further, because they already know what the reviewers will write.
This is also where the density of problems is highest. In our own dataset of analysed manuscripts, 62% of the detected weaknesses live in method and analysis, and these are the weaknesses that cannot be rewritten away later. A reviewer forgives a flat discussion. An unacknowledged design problem becomes Major revisions at best.
If you write from China, Iran, Turkey, Brazil, India or anywhere in Latin America, this is the item you cannot fully audit yourself, and it is the one that quietly costs the most. The bar is not native fluency and it is not the absence of an accent in your prose. The bar is simpler and harsher: the editor should never have to read a sentence twice to work out what you did.
That is why a paper can be scientifically stronger than the one published next to it and still get stopped at the desk. The editor is not judging your English as language, they are estimating how much work your manuscript will create for reviewers who are already saying no to three invitations a week. Effortful prose in the abstract is read as a forecast of effortful reviewing.
You can catch most of it with mechanical passes rather than intuition. Read the abstract out loud and mark every sentence where you run out of breath: those are the ones to split. Read only the first sentence of each paragraph in the introduction and check that the sequence still tells a coherent story. Search for every this and it that does not have an unambiguous noun in front of it, and for noun stacks of four or more words, which are the most common source of unreadable method sentences.
Professional language editing usually costs somewhere between 100 and 400 euros for a standard research article, depending on length and turnaround. Compare that to the real price of a desk rejection: three weeks of waiting, a resubmission cycle, and often three to six months added to the publication date of work you finished last year. If English is not your first language and the paper matters for a grant, a tenure file or a thesis deadline, that is one of the few clearly positive returns in the whole submission process. If your budget is zero, prioritise the title, abstract and first page: they carry almost all of the desk-level risk.
Write the cover letter even when the journal calls it optional. It is your only chance to speak directly to the editor before they decide, and it should answer the single question they actually have: why this paper, in this journal, now. Two tight paragraphs that connect your contribution to recent articles in that same journal do more for you than any template you can download, and a letter that simply paraphrases the abstract does nothing at all.
Then there is the layer of checks that happens before a human reads anything. Nearly every indexed journal runs submissions through iThenticate or a similar tool the moment they arrive. High overlap triggers a desk rejection automatically, and overlap with your own previously published work counts: reusing your own method section verbatim is text recycling, and paraphrasing it lightly is not enough. Rewrite it.
Everything above is verifiable against a list. This final pass is not: it requires switching sides. Print the manuscript, leave it alone for a few days if you can, then read it as someone with too little time and no affection at all for your effort. Your job for one hour is to build the strongest possible case against your own paper, in writing.
These are the questions that appear, in different words, in most real reviewer reports. Answer them on paper, in full sentences, not in your head.
The rule is simple. If you have a written, defensible answer to all eight, submit today. If two or more get a half answer, you have not finished the manuscript, you have finished a draft, and the missing week is cheaper now than the missing four months later. You are going to answer these questions either way: the only choice is whether you answer them at your desk this week or in a response letter next spring.
A desk rejection usually arrives within one to four weeks, and often within the first few days. If the paper goes out to review, the first decision typically takes two to five months, with around three months a reasonable median in psychology and health. If the journal publishes average handling times and yours is clearly past them, a short polite enquiry to the editorial office is appropriate.
Scope. Editors consistently report journal fit as the top reason, ahead of methodological weakness, formatting failures and language. That is why fit sits at the top of this checklist: it is both the most frequent cause and the cheapest to fix, because fixing it means choosing a different journal, not rewriting the paper.
Yes. It costs you an hour and it is the only direct conversation you get with the editor before the decision. A good letter does not summarise the abstract: it explains what the study adds, why it belongs in that specific journal, and confirms originality and ethics. Given that a desk rejection costs weeks, the trade is obvious.
If English is not your first language and the paper is important for a grant, a job application or a thesis deadline, usually yes. Standard editing runs roughly 100 to 400 euros and removes one of the few desk-level risks you cannot audit yourself. If that is out of reach, spend your effort on the title, abstract and first method page, which is where the editor forms the judgement.
No. Simultaneous submission is prohibited by essentially every journal and treated as misconduct. Duplicate detection systems identify it, and it can result in rejection at both venues plus a note in your editorial record. Build an ordered list of three or four target journals instead, and move to the next one only after a decision.
Think in types of reading rather than number of rounds: one content pass against this checklist, one formatting pass against the journal guidelines, and one hostile pass with the eight questions above. Leaving the manuscript untouched for a week between the content pass and the final read catches errors that no fresh reading ever will.