You have a finished manuscript, a target journal and an open submission portal. Before you upload anything, it helps to know that the biggest threat to that paper is not a hostile reviewer. It is an editor who never sends it to reviewers at all. Desk rejection, the decision to reject without peer review, is the most common negative outcome at high impact journals, and it usually happens within the first five minutes of someone opening your PDF.
The good part is that almost everything an editor checks in those five minutes is under your control the week before submission. Scope fit, the title, the abstract, the visible parts of your method, formatting compliance and the cover letter: none of that requires new participants, new data or a new analysis. It requires a few focused hours and a checklist you actually run.
This page is that checklist. It walks through what the editor reads and in what order, gives you a concrete test for each step, and ends with a two hour routine you can run on the Monday before you submit. By the end you will either send the paper with confidence or know exactly which journal you should be sending it to instead.
An editor is not a reviewer. A handling editor at a Q1 journal may receive dozens of submissions a week and process them around a full academic job. Nobody in that position reads a new manuscript from first line to last. The realistic sequence is: title, abstract, a quick scan of the participants and analysis paragraphs, a glance at the cover letter, and a check of the compliance flags the editorial office already raised (similarity report, missing ethics statement, wrong file format). For most manuscripts the decision is taken before the introduction has been read.
That ordering tells you where to spend your last week. A beautifully argued discussion cannot rescue a title that signals the wrong journal, and a thorough limitations section cannot rescue an abstract that hides what you found. Effort spent on the first 300 words of your submission package returns far more than effort spent anywhere else.
There is a second thing worth internalising. The editor is not asking whether your science is good, but whether two busy, unpaid reviewers will spend six hours on your paper and come back with something publishable. Every check below is that same question from a different angle: does this manuscript make the editor's decision easier or riskier?
Scope mismatch is the most frequently cited reason for rejection without review, and it is the most expensive one, because no rewriting fixes it after the letter arrives. It is also the check most authors skip, because it feels like it was already settled when they picked the journal months ago.
Being in scope does not mean your topic sounds similar to what the journal publishes. It means the journal has recently published papers with your kind of question, your kind of design and your kind of sample. Three tests settle that in under half an hour, before you write a single new sentence.
Open your own reference list and count how often each journal appears. If a journal shows up five or six times, that is where your academic conversation is already happening and it is a natural candidate. If your target journal appears zero times, treat that as a warning sign you can see for free: you are proposing to join a conversation you have not been reading. Either you have missed relevant work from that journal (fix the introduction) or the journal is not the right home for the paper (fix the choice).
Open the tables of contents of the last four to six issues and look for at least two articles comparable to yours in design and population. Not comparable in topic: comparable in design. A cross sectional questionnaire study sent to a journal that has published only randomised trials and longitudinal cohorts for two years is out of scope however well the topic matches. If you cannot find those two papers, the journal title is promising something the journal does not actually do.
Read the aims and scope statement all the way to the last line, looking specifically for what the journal refuses. Many journals state explicit exclusions: no pilot studies, no single site convenience samples, no scale validations, no student samples, no purely descriptive reports. These sentences usually sit in the final paragraph, which is exactly the paragraph most authors stop reading before. Ten minutes there can save you three months.
Your title has one job: let the editor place the paper correctly in about four seconds. A working title states the population, the design and the outcome, and drops the decorative scaffolding. Compare “An exploratory study on the relationship between certain psychological variables and academic performance” with “Academic burnout and grade point average in 812 first year psychology students: a cross sectional study”. The second one tells the editor what you did, with whom, and how many. It also protects you, because the design is stated up front rather than discovered later as a disappointment.
Watch your verbs. Causal language in the title of a correlational study (“increases”, “leads to”, “improves”, “the effect of” applied to a single time point survey) reads to an experienced editor as a signal about the rest of the paper. If your design cannot support the verb, use “associated with” and keep the stronger claim for a study that earns it.
The abstract has to survive alone, because in the editorial filter it is alone. It needs the gap in one sentence, the design named explicitly, the sample with its actual n, the main results with effect sizes and confidence intervals rather than p values by themselves, and a closing implication that says something specific. A useful test: send only the abstract to a colleague from a neighbouring field and ask them what you found and how. If they have to open the PDF to answer, the abstract failed the same filter the editor applies.
If English is not your first language, the bar is not perfect idiom. The bar is that the editor never has to read a sentence twice to work out what you did. Two cheap fixes: read the abstract aloud and split in two every sentence you stumble on, and pay for professional language editing on those 250 words specifically. That is where the money buys the most.
Experienced editors diagnose methodological problems straight from the abstract. A structural equation model on 60 participants. A mediation model estimated from a single time point. Causal conclusions from a cross sectional design. Results reported as p values with no effect sizes. Psychometric instruments used with no reliability reported in this sample. None of these require reading the method section, and each of them predicts what reviewers will say, which is precisely the risk the editor is trying to avoid.
So make the defensible version visible early. Justify your sample size in one honest sentence: an a priori power analysis, or a frank statement of the practical constraint plus a sensitivity analysis showing the smallest effect you could detect. Report effect sizes with intervals next to every headline result. If a measure has weak reliability in your sample, say so before a reviewer does, and say what you did about it.
Our own numbers point at the same territory. Of 346 manuscripts analysed with the paper reviewer tool, 62% of the detected weaknesses sat in the method and analysis sections, and 84% received a Major revisions verdict overall. That is the same zone an editor scans first, for the same reason: it is where reviewers concentrate their objections, and the editor is trying to predict the reviewers.
This category stings, because it has nothing to do with your science and it still ends submissions every week. Being 40% over the word limit, ignoring the required section structure, submitting references in the style of the previous journal you tried, uploading figures at screen resolution or forgetting the anonymised version for double blind review: each of these tells the editor you did not read the instructions. The inference that follows is the damaging one, that you will also be hard work at the revision stage.
Then there is the statement block, which is now mandatory almost everywhere and is a common trigger for automatic return or rejection. Before you submit, confirm you have all of these ready: ethics committee approval with the institution name and the approval reference number, informed consent, funding, conflicts of interest, data availability, author contributions, trial or preregistration number when it applies, and the reporting guideline checklist your study type requires (CONSORT for trials, STROBE for observational studies, PRISMA for reviews, COREQ for qualitative work). Several journals desk reject submissions that arrive without the completed checklist, without ever looking at the content.
Finally, remember that the editorial office runs a similarity check before a human reads anything. Reused text from your own earlier papers counts, and the method section is where authors reuse most. Light paraphrasing does not solve it: rewrite the passage properly. A high similarity score reaches the editor as a flag attached to your file, and no editor wants that conversation.
Most cover letters are wasted because they answer the wrong question. The editor is not asking what your paper is about (that is what the abstract is for). The editor is asking why this paper, in this journal, now. Three short paragraphs handle it. First, what you did and what you found, in two sentences, with the actual numbers. Second, why it belongs in this journal, referring to one or two specific papers the journal published in the last couple of years and saying how yours extends, challenges or complements them. Third, the compliance sentences: original work, not under consideration elsewhere, all authors approved the submission, ethics approval obtained, conflicts declared.
That second paragraph is the whole letter. It is the only part that must change for every journal, and the part a generic template cannot fake. An editor who reads a specific, accurate sentence about their own journal knows immediately that this is not a shot fired at fifteen titles in a row.
Things to leave out: compliments about the journal's prestige, the abstract copied and pasted, and the claim that your study is the first of its kind unless you can name what it is first relative to. If the journal asks for suggested reviewers, supply them properly: real people who work on your topic, no co authors from the last five years, nobody from your institution, with institutional email addresses. Leaving that field empty slows the process and at some journals gets the submission returned before it is assigned.
Everything above collapses into one working session. Block two hours, close your email, and go through it in this order, because each step can save you from doing the next one on the wrong journal.
If it fails on scope, change the journal now rather than after two weeks of waiting for a three sentence letter. Nothing you write this week will fix a mismatch, and the next journal on your list probably needs only a reformatting pass. If it fails on the abstract, the visible method or the statement block, delay the submission by a week and repair it. The arithmetic is not close: a week of work against three months of queue, a reformatting cycle and a decision that tells you nothing about your science.
Keep the risk asymmetry in mind when you decide how high to aim. A desk rejection returns your manuscript in one or two weeks, so an ambitious journal is a cheap bet, provided your paper cannot fall for the avoidable reasons: formatting, missing statements, an unreadable abstract or a similarity flag. Skip those and the bet is just a slow way of losing a month.
It depends on the journal's volume and selectivity. High impact generalist journals commonly reject more than half of submissions without review, and figures reported for specialised psychology and health journals typically fall between 20% and 40%. The pattern is consistent: the lower the overall acceptance rate, the larger the share of rejections that happen before peer review starts.
Normally between 48 hours and three weeks. If a decision lands in the first week, the manuscript almost certainly never left the editor's desk. A decision that takes three months or more usually means peer review did happen, even when the letter arrives without attached reports.
It matters as a signal, not as an aesthetic judgement. An editor cannot verify your analyses in five minutes, so they use proxies for how carefully the work was done, and compliance with clearly published instructions is the cheapest proxy available. Missing ethics statements or a missing reporting checklist go further than a signal: at many journals they trigger an automatic return or rejection before the editor reads anything.
Not by itself. Editors are used to reading English written by researchers all over the world, and imperfect idiom is not the threshold. The threshold is comprehension: if the editor has to read a sentence twice to work out what you measured, they assume reviewers will refuse to do that work. Concentrate your effort (and any editing budget) on the title, the abstract and the first paragraph of the method, which is what actually gets read in the filter.
You can reuse the structure and the compliance sentences, but the fit paragraph has to be rewritten every time, because it is the only part that does any work. A letter that names two recent papers from the journal you are submitting to reads completely differently from one that praises the journal's reputation in general terms, and editors recognise the difference immediately.
Often yes, provided the avoidable failures are closed first. A desk rejection costs you one or two weeks, which is a small price for a chance at a better outcome, so the bet is reasonable when scope fit is real and formatting, statements, abstract clarity and similarity are all clean. It stops being reasonable when you are gambling on problems you could have fixed in an afternoon.