Journal Submission Checklist Before You Hit Submit

The manuscript has been finished for a week and the submission form still is not. Most of the remaining time goes on documents nobody drafts while writing a paper: a data availability statement, a funding statement, a competing interests declaration, and a list of who did what.

A journal submission checklist is most useful when it follows the order the submission system asks for things, because that is the order in which a missing item stops you. What follows is that order, from the manuscript file to the confirmation screen, with the thing most likely to go wrong at each step.

A formatting miss is expensive here: Elsevier's own author guidance names format non-compliance and "sloppy copy" among the reasons a manuscript is turned down before review, and Springer Nature lists poor structure and poor presentation on its own account. What happens to a submission once it clears this screen is a separate subject.

The Journal Submission Checklist, in the Order the Portal Asks

The submission system gathers things in a fixed order: the manuscript file, then the declarations, then the file uploads, then the cover letter. The secret here's having each answer ready when its field appears. Anything invented inside the box gets written under time pressure and rarely revisited.

Item	What the form expects	The common miss
Manuscript file	The stated file type, with line and page numbers if required	Author names left in the file for a double anonymized journal
Title and short title	Both, inside the journal's own character limits	A short title written to fit a different journal's cap
Abstract	The journal's word limit, and headings only if it asks for them	A structured abstract pasted into a journal that wants one paragraph
Keywords	The number the journal asks for, in its own field	Words repeated from the title, which adds no new route to the paper
Authors and affiliations	Full names, order, affiliations, one corresponding author	An author order agreed in conversation and never confirmed in writing
ORCID	A registered iD for the corresponding author, often for all authors	A coauthor registering for the first time while the form is open
Funding	Every funder, with grant numbers exactly as awarded	A grant number transcribed from memory
Competing interests	A statement even when there are none, in the journal's wording	An empty field instead of a declaration that none exist
Data availability	Where the data are, or a stated reason they are not available	Data on request, with no named contact and no repository
Author contributions	Who did what, often against a fixed taxonomy of roles	Roles assigned at submission, with one coauthor unreachable
Ethics and consent	Approval body, reference number, and consent wording where relevant	The approval number sitting in a supervisor's files
Figures and tables	Separate figure files at the stated resolution, numbered and captioned	Figures embedded in the manuscript when the journal wants them apart
Supplementary files	Labelled, and referred to by name in the manuscript text	A supplementary file the paper never mentions
Suggested reviewers	Names, affiliations and institutional emails, conflicts excluded	A recent coauthor, who is then disqualified
Cover letter	Addressed to a named editor, naming this journal	The previous journal's name still in the file

Manuscript, Figures, Tables and Supplementary Files

This is the one thing that the system is more likely to reject than a person would. File format: journals say what file type they accept. Many like continuous line and page numbers so a reviewer can point at a specific line. Double anonymized journals want identifying details stripped out of the file and the document properties (where author names are stored).

Two things inside the manuscript are worth checking against the target journal's own page rather than the last journal you submitted to: the abstract word limit, and whether the abstract takes headings. IEEE journals require one paragraph of 150 to 250 words with no numbered equations, reference citations or footnotes; PLOS ONE allows 300 words and asks for no headings. Writing an abstract for a research paper that already fits the target's rules is quicker than rebuilding one that does not.

Figures usually upload as separate files at a stated minimum resolution rather than as images inside the manuscript, and the text then carries a caption and a callout for each one. Tables belong in the manuscript in editable form, not as screenshots. Supplementary files need names that match how the paper refers to them, since a file called appendix_final tells a reader nothing.

Word limits apply to more than the manuscript: the abstract, the short title, and at some journals a significance statement or a plain language summary, each with its own cap. Checking each against the journal's stated number beats discovering the overflow inside a portal field that truncates silently. A set of free writing tools covering word counts, title case and reference formatting turns that pass into a few minutes.

The Statements Nobody Drafts Until the Form Asks

Four statements are standard across major journals and none of them gets written while the paper is being written: competing interests, funding, data availability, and author contributions. Each takes minutes to draft and hours to chase, because the information sits with other people.

A competing interests statement is required whether or not any exist, and an empty field reads as an omission rather than a declaration that there are none. Funding statements want every funder named with grant numbers exactly as awarded, worth copying from the award letter rather than from a coauthor's memory. Both usually appear in the manuscript as well as the form, so the file and the fields have to agree.

A data availability statement says where the data are and how a reader reaches them: a repository name with an accession number, a stated restriction with the reason for it, or a clear statement that no new data were generated. Some journals accept data available on request from the corresponding author and others require a deposit, so check which one the target is before writing that sentence.

Author contributions are recorded against a fixed taxonomy at many journals, with each named author mapped to roles such as conceptualization, methodology, analysis or writing. Underneath sits the harder question of who qualifies as an author at all. COPE's authorship guidance rests on two minimum requirements, a substantial contribution to the work and accountability for it, and aligns with the ICMJE four-part test: substantial contribution to conception, design, acquisition, analysis or interpretation; drafting or critically revising the work; approving the final version; and agreeing to be accountable for all of it. Contributors who meet fewer than all four are acknowledged instead, and funding acquisition, general supervision or language editing alone qualifies nobody.

ORCID, Reviewers and the Items That Depend on Other People

Many journals now need an ORCID iD for at least the corresponding author, and some need one for every author on the paper. Before opening the submission form, register and link it. The field appears early, and a coauthor who has never registered becomes a blocker at the worst moment.

Suggested reviewers are optional at some journals and mandatory at others, and the useful list is researchers who have published on the same question with no tie to you. Recent coauthors, colleagues at your own institution and anyone who has already read the manuscript are normally excluded. The names go in the form's own field: Elsevier's guidance is explicit that a cover letter should not include "funding information, author declarations, or suggested or opposed reviewers".

The Cover Letter Is One Item on the List

It's the last field on most submission forms, and also the only one, which involves writing (as opposed to data entry). The cover letter introduces the paper to an identified editor, explains what's new about it, and explains why this journal. There are two possible failures here, both mechanical.

The first is the previous journal's name surviving in the file. Springer Nature's cover letter guidance asks for a confirmation sentence that names the target directly: "We confirm that this manuscript has not been published elsewhere and is not under consideration by another journal. All authors have approved the manuscript and agree with its submission to [insert the name of the target journal]." The placeholder exists because the sentence gets reused, and it carries the last journal's name until somebody changes it.

The second is padding the letter with what the form already collected. What belongs there instead is close to Elsevier's own list: the manuscript's category, a short statement of its central theme, and its novelty and broader implications for the field. A journal cover letter generator that starts from the paper's title, abstract and target journal removes the blank-page part, the bit that gets left until last.

Avoiding desk rejection is what the checklist is for, and responding to reviewer comments is the step that follows if the paper clears it.

Run the whole list once with the portal open and nothing filled in, so you find out what is missing while there is still time to ask somebody for it. The item that stops a submission is almost never the manuscript. It is the ethics approval number in a supervisor's filing cabinet, and the supervisor is on leave.