November 16, 2026

How to Run a Manuscript Pre-Submission Check

A pre-submission check is a timed sequence of decisions, not a final proofread. You are answering one question at the end: if a reviewer reads this today, will they find the manuscript self-contained, consistent, and ali

Written byWisPaper TeamAI Research Workflow Team
Editorial illustration for How to Run a Manuscript Pre-Submission Check

A pre-submission check is a timed sequence of decisions, not a final proofread. You are answering one question at the end: if a reviewer reads this today, will they find the manuscript self-contained, consistent, and aligned with the journal's expected format? Most rejection-by-desk comes from mismatched structure, incomplete reporting, or a gap between the methods and the figures. The check therefore has three layers: submission requirements, reporting completeness, and internal consistency. Run them in that order. The first catches avoidable errors; the second catches evaluative gaps; the third catches the errors a reviewer notices before the results.

This guide gives you a complete sequence for a thesis chapter or first manuscript, with checkpoints after each layer and a worked example using a mixed-methods study being prepared for a January submission. The final result is not a perfect paper. It is a paper where every remaining decision is explicit and every journal-facing requirement is met.

Which starting state are you in?

Start by naming the manuscript type. The check is different for a coursework paper, a thesis chapter being converted, and a first submission of completed research. Do not apply all sections to all cases.

A coursework paper usually needs to satisfy a rubric and a word limit before it needs journal formatting. A thesis chapter often carries dissertation conventions such as chapter numbering, signposting, and literature-review framing that a journal manuscript must remove or compress. A first journal submission needs the journal's current instructions applied as its controlling document. The International Committee of Medical Journal Editors frames this directly: authors should follow the selected journal's current instructions for article type, format, required statements, and submission materials. If you are submitting to a biomedical venue, treat those instructions as the first artifact you collect.

Write a one-line starting state: manuscript type, target venue or grader, current word count, and the one thing you most suspect is wrong. For example: "Original research article for Journal of Mixed Methods Research, 5,900 words, worried the qualitative analysis is not described." That suspicion becomes the first item on your consistency layer, but do not run there yet.

What is the controlling document for the check?

The controlling document is the journal's instructions or your course rubric. Download it and make a checklist before opening the manuscript. If the journal follows ICMJE-style preparation, it will specify article type, format, statements, and the exact submission materials. Some journals also name a reporting guideline.

Choose a reporting guideline only after the article type is fixed. The EQUATOR Network database organizes reporting guidelines by study type and specialty, so you can match a checklist to the design you actually ran, not the design you wish you had run. For a mixed-methods study, the relevant standard may come from a design-specific module rather than a single generic list. The American Psychological Association's JARS standards similarly structure reporting by study design, with separate modules for quantitative, qualitative, and mixed-methods research so readers can evaluate how the work was conducted and analyzed.

The checkpoint is simple: you can name the controlling document, the reporting guideline, and the specific checklist module before you edit anything. If you cannot, the first fix is to acquire the checklist, not to improve the prose.

Which reporting gaps does the checklist expose?

Work through the checklist against the manuscript as a table. Create four columns: checklist item, manuscript location, present or missing, and action. Do not rewrite during this pass. You are producing a gap map.

The gap map creates the first real segmentation of the work. Items marked missing become the revision order for reporting completeness. Items marked present but thin become the second priority. Items you cannot locate become a special class, because "I am sure I described that" is exactly where reviewers stop reading.

For a mixed-methods manuscript, the design-specific module often exposes three common gaps: the rationale for mixing methods is implied rather than stated, the point of integration is described as a result rather than designed, and the order of quantitative and qualitative phases is ambiguous. A coursework paper may show a different gap profile: a methods section written as a summary of the course materials rather than a record of what was actually done.

Keep the gap map. You will return to it as the scoring sheet for completion.

A worked example: the winter-break mixed-methods manuscript

Set the example concretely because the decisions become visible. A doctoral student finishes data collection in December. They have a quantitative survey of 300 participants and 18 follow-up interviews. They plan to submit to a journal that requires CONSORT-style trial reporting for clinical designs but a mixed-methods module for this study. They run the pre-submission check over three days of winter break, not one uninterrupted session.

Day one: collect the journal instructions. The student identifies that the journal requires a separate title page, a structured abstract with a registered report statement, and a data availability statement. Two of these are missing from the thesis chapter draft. The student also opens the EQUATOR Network database, confirms the study type, and downloads the relevant reporting checklist module. They now have a gap map of 23 items, of which 8 are missing and 15 are present but inconsistent in placement.

Day two: reporting completeness. The student fills the eight missing items. Three are structural, such as adding the data availability statement. Five are evaluative, including a statement of the point at which the quantitative and qualitative data were integrated. The student realizes the integration point was not designed before data collection, which is a true limitation. The fix is not to invent a rationale. It is to state exactly what was done and move the integration discussion out of the results section into the methods. That is an honest manuscript improvement, not a cosmetic one.

Day three: internal consistency. The abstract claims 312 survey responses, but the methods table reports 300 after exclusions. The interview count says 18, while the participant table shows 18 for interviews but 17 for informed-consent records. The student reconciles each number to a single source: the cleaned dataset. This is the layer that decides the reviewer's first impression.

The student's completion decision is specific: every number in the abstract, methods, results, tables, and figures now matches one stated analysis dataset, and the checklist gap map has no open items. The paper still has a real methodological limitation, but it is now disclosed and positioned correctly.

What does the submission requirements layer produce?

This layer produces the submission packet, not the manuscript quality judgment. Run it first because it is fast and because a missing component stalls the entire submission. The artifact is a file list checked against the journal instruction list.

Typical components for a journal manuscript include the manuscript text with line numbers, a separate title page, a cover letter, figures at the requested resolution, tables in the requested format, a disclosure or conflict of interest statement, author contribution statements if required, a data availability statement, and any reporting checklist file. The cover letter should not restate the abstract. It should identify the article type and the main decision the paper enables for the journal's readers.

The checkpoint for this layer is a zero-gap condition. For every required component, there is a physical file or a filled section in the submission system. If one component is "to be completed later," the layer is not done. The discrete action is usually small, but the discipline of producing the file now removes the illusion of progress.

How do you run the internal consistency pass without rereading linearly?

A linear reread does not catch the errors that matter. A reviewer will sample from the paper in a different order: abstract, first figure, methods participant numbers, results first sentence. Run your consistency pass in that order.

Start by extracting every number in the abstract, results, tables, and figures into one column. The rule is that each number traces to exactly one stated denominator, and all denominators trace to one analysis dataset. A number without a denominator is a consistency defect. A percentage with no total, a mean with no sample size, a count with no inclusion criterion: these are the errors that feel small and read as careless.

Next, read the abstract and the discussion side by side. The abstract's conclusion must be the discussion's conclusion. If the abstract says "mixed methods improved interpretation" and the discussion says "the quantitative data provided context for the qualitative themes," you have a machinery problem: two versions of the same finding. Choose the version that matches the methods section and revise the other.

Finally, check the tense and voice consistency of the methods. A methods section that shifts from past tense for data collection to present tense for analysis reads as though the analysis was not finished when the section was written. Standardize to past tense for everything the authors did and present tense only for well-established instruments or analytical choices still true in the manuscript's present tense reading.

When is a manuscript actually ready?

Readiness is not confidence. It is the presence of four closure conditions: the checklist gap map is empty, the consistency pass has produced a single clean data trace, the submission packet is physically present, and every remaining weakness is either fixed or explicitly placed as a limitation. If any of these four is missing, the manuscript is not ready; it is merely formatted.

This means the check can end in a decision to revise for one more week because the gap map exposed a methods description that is genuinely thinner than what was actually done. That is a successful outcome of the check. It is the difference between a paper that is ready in form and one that is ready for evaluation.

You will know the result is ready when you can answer one question without hedging: if a reviewer asks "what did you do and how," can the manuscript answer from its own sections without the author providing an explanation outside the text? If the answer depends on an email to the editor or a footnote you have not written, the check is not finished.

A reusable pre-submission checklist

Use this as the master table. Each row is a checkpoint, not a task description, and each is either done when the stated artifact exists or not done.

LayerCheckpointDone when
Submission requirementsControlling document collectedJournal instructions or rubric saved as a file
Submission requirementsArticle type selectedArticle type matches the journal's options
Submission requirementsRequired statements draftedData availability, disclosure, contribution, ethics statements all written
Submission requirementsFile list matchedEvery required component has a physical file
Reporting completenessGuideline chosenA design-specific checklist module is open
Reporting completenessGap map completeEvery checklist item mapped to present, missing, or thin
Reporting completenessMissing items filledGap map has no open items
Internal consistencyNumber trace cleanEvery result number traces to one stated denominator
Internal consistencyConclusion alignedAbstract conclusion matches discussion conclusion
Internal consistencyMethods voice standardizedPast tense for actions, present tense for stable instruments
CompletionFour conditions metGap map empty, data trace clean, packet present, limitations explicit

Do not skip a row by arguing the journal "probably doesn't require it." The row exists because the instruction either lists the item or leaves the choice to the author. If the instruction leaves a choice, the done condition is a written decision, not an omission.

How should you organize the research materials for the next revision?

A pre-submission check usually ends by generating a small set of next-step files: the revised manuscript, the filled reporting checklist, the reconciled dataset or a cleaned excerpt, and the list of remaining limitations. The next manuscript, whether it is a revision of this one or the next project, runs faster if those files are stored as a named project rather than a folder of emails and downloads. Organize the project by the artifact type that generated the revision, not by the date you last touched the file. A project structure that mirrors the checklist layers, submission, reporting, consistency, and limitation, lets you reopen the exact check that is still open.

If you are using WisPaper as your research workspace, you can create a new research project to hold the manuscript source and the checklist files together. As you move toward the next paper, Idea Discovery can take the topic, background, references, and constraints you have gathered and generate research ideas for the follow-on work. The point is not to automate the next manuscript. It is to make the boundary between one submission and the next explicit, so the check artifacts do not evaporate.

Removing citation risk before the next submission

The final pre-submission risk is not in the manuscript's prose. It is in the reference list. A citation that does not resolve to a real paper damages the submission independent of every other layer, because the editor or reviewer checks references before accepting the argument. The fix is mechanical and should happen after the reporting and consistency layers, not alongside them. Export the reference list as BibTeX, and screen each entry against a real academic database rather than against your memory. TrueCite checks BibTeX against real academic databases to flag references that do not correspond to a published work. If you have been gathering sources during the project, My Library stores and organizes the papers you have uploaded, so the check runs against the same set you actually read rather than a list reconstructed from memory.

The completion criterion for this layer is zero flagged entries that you cannot resolve to a real publication. If one entry is flagged and you cannot find the paper in any database, remove it or replace it with a source you have actually read. This is a deletion decision, not a formatting decision, and it is part of submission readiness.

FAQs

The check is timed by the gap map, not by the clock. For a first journal manuscript, expect the submission requirements layer to take under one working session, the reporting completeness layer to take one focused day, and the consistency pass to take one more focused day, with gaps between sessions. A thesis chapter conversion may take two additional sessions because structural cuts are decisions, not edits. If the check takes an hour, you probably ran only the consistency pass and skipped the checklist, which means you checked readability rather than submission readiness.