Certification Procedure

1. Scope and status of this document

1.1 · What this Procedure states. This Procedure states how an assessment is carried out on XSci: what happens at each stage, who does it, what is recorded, and what each party may and may not do. It is the operative companion to two other documents:

(a) The Certification Standard states what is assessed — the criteria, the findings, and the conditions of refusal. (b) The Peer Review Policy states what review on XSci decides and what certification means. (c) This Procedure states how the assessment is conducted.

1.2 · Why this document is binding rather than descriptive. Two of the four grounds of appeal turn directly on it: that the Certification Procedure was not followed, and that a report does not meet the standard of detail this Procedure requires. An appeal on either ground is measured against the text below. It is published for that reason, free to read and without an account, so that an author considering an appeal can see exactly what they are entitled to.

1.3 · Where this Procedure sits. It may not vary anything in the Certification Standard or the Peer Review Policy; where it appears to, those documents govern. It is not a substitute for the Guide for Reviewers, which explains the same process to reviewers in the second person and carries no contractual force of its own.

1.4 · Must and may. Where this Procedure uses must, the obligation is binding on the person or party named. Where it uses may, a genuine discretion is intended.

1.5 · Two amounts, named consistently. The assessment fee is what an author pays XSci for an assessment. The reviewer payment is what XSci pays the reviewer who carries it out. Neither is calculated from the other.

2. Before an assessment exists

2.1 The deposit

2.1.1 · An assessment is always bought against a deposited version. Depositing is free, is never conditional on any purchase, and results in a dated record with its own DOI. XSci does not assess unposted work. Because the DOI exists before an assessment can be bought, no step in this Procedure can affect whether a manuscript is available or citable.

2.1.2 · Depositing is not peer review, decides nothing about quality, and is not a review outcome of any kind. Nothing in this Procedure begins until an assessment is bought.

2.2 What the purchaser fixes at purchase

2.2.1 · Three things are fixed at the moment of purchase, and not afterwards.

2.2.2 · The version. The assessment is of the version identified at purchase, and of no other. The author's one response comes after the decision and is commentary on the record (clause 8); a revised deposit is a different version and requires a new assessment of that version.

2.2.3 · The target venue. Up to three named journals, or a blank which resolves to the default the top journals in this field. The target is fixed at purchase, and no reviewer may make certification conditional on the authors renaming it.

2.2.4 · The disclosure election, and therefore the price. There are two prices and the disclosed price is the lower. A report bought at the discounted price is published automatically 48 hours after the report is released to the author, with no further step by the author and no confirmation stage. This is stated wherever the discounted price is offered.

2.3 Changing the disclosure election afterwards

2.3.1 · Withdrawing disclosure costs the difference between the two prices, is charged once only per assessment, and is calculated at the prices in force on the date of purchase.

2.3.2 · Opening a privately bought report is free at any time and is not refunded.

2.3.3 · Neither election has a closing date. A privately bought report may be opened by the author years later, without the reviewer's agreement being sought. Reviewers are told this at assignment and must write every report on the basis that it may become public.

3. Selecting the reviewer

3.1 Where reviewers come from

3.1.1 · The candidate pool. Candidates are drawn from the authorship of closely related published work, and from among researchers who have published in the target venue named for the assessment, with a preference for the latter where a suitable candidate is available. Having published in the target journal is a qualification, not a conflict.

3.2 How a candidate is chosen

3.2.1 · Matching. Candidates are matched on subject-matter expertise and on the absence of any barred interest. XSci may use internal tools to assist in identifying candidates, but assignment is never determined solely by an automated system, and no automated system certifies anything, alone or in combination with any other.

3.3 The eligibility check, before acceptance

3.3.1 · Four bars. A candidate must decline outright if any of the following applies. Each is a bar, and disclosing it does not cure it:

(a) a prior role in this work — having advised on it, reviewed it for a journal, or taken part in producing or correcting any part of it; (b) a close personal relationship, or a dispute, with any author, of a kind a reasonable person would think material; (c) a financial or commercial interest in the outcome, or in the subject matter to the extent the outcome would affect it; (d) a current supervisory relationship with any author, in either direction.

3.3.2 · One exception. A reviewer who assessed an earlier, certified version of the same manuscript is not barred by the prior-role bar in 3.3.1(a) from a later version.

3.3.3 · Everything else is declared, not barred — recent co-authorship, shared affiliation, competing work, academic rivalry, a strongly held prior public position. The candidate declares it, XSci decides whether to proceed, and the declaration goes on the assessment record.

3.3.4 · A candidate who cannot judge an area the manuscript engages must say so before starting. XSci then changes or adds a reviewer. Such an area must not be recorded as not applicable.

4. The assignment

4.1 · The clock starts on acceptance. XSci records the assignment. The reviewer is told at this point: the version, the target venue as fixed at purchase, the disclosure election in force, and the version of the Certification Standard and of this Procedure that apply.

4.2 · Delivery and the reviewer payment. A report is delivered within five to ten days of assignment in the ordinary case. The reviewer payment is the full published rate for a report delivered within 5 days, half from 5 to 10 days, and after day 10 the assignment is reissued to another reviewer — the original reviewer may still submit, but unpaid. The published rate does not vary by country and cannot be reduced for an individual reviewer.

4.3 · Neither the assessment fee nor the reviewer payment varies with the decision. No part of a reviewer payment depends on the outcome.

4.4 · A bar discovered after acceptance. A reviewer who discovers a barred interest partway through must stop and tell XSci immediately. The assignment is reassigned. There is no consequence to the reviewer for doing so, at any point.

5. Conducting the assessment

5.1 The two questions

5.1.1 · Both limbs must hold, and they are independent.

(a) No major error. A major error is a defect which, while it stands, means the central claims are wrong, unsupported, or impossible to evaluate — as distinct from something the authors could reasonably be asked to put right. Anything in the second class is ordinary feedback and does not stand in the way. (b) A genuine prospect of publication in the target venue, assuming revision of the kind a journal would normally ask for.

5.1.2 · A manuscript may be entirely sound and still not certified, because it was aimed higher than it reaches. That is an ordinary outcome, not a criticism, and it requires no major error anywhere in the report.

5.2 The nine domains, and the four findings

5.2.1 · One finding and one sentence per domain. For each of the nine assessment domains set out in the Certification Standard, and for each domain the manuscript engages, the report records exactly one of four findings, together with at least one sentence specific to that manuscript saying what was looked at and what was found:

no concern · concern noted but not material · material concern · not applicable, with a reason

5.2.2 · Nothing is counted. The findings must not be summed, averaged, expressed as a fraction, converted to a number or a letter, or used to sort or rank anything, by XSci or by anyone downstream. They are never displayed as the outcome. They exist so that every domain receives deliberate attention.

5.2.3 · Not applicable describes the manuscript, not the reviewer. A domain is not applicable only where the manuscript does not engage it at all. It is never used because the reviewer lacks the expertise, the materials or the time to reach a view: a reviewer who cannot assess a domain the manuscript does engage must tell XSci, and must not record it as not applicable. Recorded properly, not applicable says nothing about the quality of the work and is never a mark against it.

5.2.4 · This part of the report is delivered in full whatever the decision. It is what the authors can act on, and it is not cut short because a view on limb 5.1.1(b) has already formed.

5.3 Verifiability: record how far you actually got

5.3.1 · The three-step ladder. Verifiability is recorded on a three-step ladder, and the report states the highest step actually reached and no further:

(a) the underlying materials are declared; (b) they are reachable and as described; (c) the reported results follow from them.

5.3.2 · What underlying materials means. Whatever the manuscript's claims rest on and which another competent person would need in order to check them — measurements, data, code, transcripts, field notes, source texts, images, instrument or archival records, derivations and proofs, protocols, specimens. No kind of material is treated as better evidence than another, and a manuscript with no data and no code is not less verifiable for it.

5.3.3 · Absence is recorded as absence, not as a fault. Where materials were not provided, the report says what was missing.

5.3.4 · Replication is outside the ladder. Whether findings would hold up on new data is never assessed and never implied. No report comments on it.

5.4 Judging the target venue

5.4.1 · Named and default targets. Where journals were named, the reviewer judges against those, and potential for any one of them is enough. Where the target is the default, the reviewer decides what counts as a top journal in the field and must name in the report which journals they had in mind — a default target left unnamed is not an acceptable answer.

5.4.2 · The judgement is on the tier, not on one editor's taste. It is not a prediction of an editorial decision.

5.4.3 · A reviewer who cannot place the work must say so. That is a refusal with that stated as the reason, and it is never a ground to certify by default.

5.4.4 · Certification must never be made conditional on the authors renaming their target.

5.5 What is outside the assessment

5.5.1 · Four things no report decides. No reviewer is asked to decide: whether the findings are true; whether they would come out the same way from new data; writing quality beyond what is needed to evaluate the work; or who the authors are, where they work, or who funded them.

6. The decision

6.1 · Two decisions, and no others. The reviewer records one of:

(a) Certify — no major error remains in any applicable domain, and the work could reach the target venue. (b) Do not certify, because — each reason stated specifically: anchored to the place in the manuscript for a major error, and stated against the target venue for a question of fit. The report says which kind each reason is, because one tells an author this needs fixing and the other this is not the right home for it.

6.2 · The decision is binary. There is no score, grade, rating, band, star or ranking, and none may be introduced.

6.3 · A reviewer may not: ask for revisions as a decision; make certification conditional on any further purchase; or contact the author.

6.4 · One version, one reviewer. A version is assessed once, by one reviewer, and no second opinion is available at any price. Where more than one reviewer is ever used, the number is fixed before any report exists, reviewers do not see one another's reports, the record states the split in words and never rounds it up, and one certifying reviewer is sufficient. The single exception is stated in clause 10.

6.5 · No member of XSci staff may ask a reviewer to change a decision, and none has the authority to do so.

7. The completeness check

7.1 · What is checked, and what is not. After the reviewer has recorded the decision and before the report reaches the author, XSci checks it for completeness, not for its conclusion. This is the check appeal ground 4 is measured against, and it covers only the following:

(a) whether every domain the manuscript engages carries a finding and at least one sentence; (b) whether the reasons given are specific rather than generic, and anchored where the Standard requires; (c) whether the judgement on the target venue is present and reasoned; (d) whether, for a default target, the reviewer has named the journals they had in mind; (e) whether the verifiability record states the step actually reached and claims no more; (f) whether the competing-interest declaration and the record of AI use are present; (g) whether anything in the report belongs in an integrity referral instead (clause 9).

7.2 · Where a report is incomplete it is returned to the reviewer, with the specific omission identified, up to three times; the assessment is then reassigned to another reviewer.

7.3 · The certification decision is never a subject of this check, and no one performing it may ask a reviewer to change a decision or has the authority to.

8. The author response

8.1 · One response per assessment. Once the report and the decision have been released, the author is entitled to one response. The response is commentary: it is recorded on the record beside the report, and it does not reopen the assessment. The decision was made on the version presented to the reviewer, and neither the response nor anything in it changes that decision, the contents of the report, or the version a certification names.

8.2 · A revised version is a new assessment. There is no revision inside an assessment. An author who wants a decision on a revised version deposits that version — free, dated and with its own DOI, as any deposit is — and buys an assessment of it on the ordinary terms and at the ordinary assessment fee.

8.3 · Reviewers and authors do not communicate directly, in either direction. All communication passes through the report and this response mechanism, under the platform's default single-blind arrangement.

8.4 · The one-response limit applies to the assessment only. It does not apply to, and is not consumed by, an integrity inquiry (clause 9), in which an author may make as many representations as the inquiry requires.

9. Integrity referrals

9.1 · When a reviewer must refer and stop. A reviewer who encounters plagiarism, undisclosed duplicate publication, or signs that images or data have been manipulated — or an undisclosed competing interest material to the findings — must refer the matter to XSci and stop, and must not write the allegation into a report that may be published. The referral states what was observed and where, facts only and no conclusion about intent.

9.2 · The assessment shows as in progress, with no public reason given. No public notice, flag or status change identifies the manuscript as the subject of an enquiry while the matter is open. A suspicion is not made public before anyone has made a finding, and this is a protection for the authors.

9.3 · The reviewer takes no part in deciding the matter.

9.4 · The authors are given an opportunity to respond before any finding is made, and may respond as often as the inquiry requires.

9.5 · Where the matter is not made out, the assessment resumes and concludes normally, and no record of the enquiry appears on the public record.

9.6 · Where it is made out, XSci acts under the Retraction and Correction Policy and the assessment is closed. Because no report is delivered, the assessment fee is refunded in full, as a case of non-delivery under clause 10 and the Peer Review Policy. XSci issues the refund itself: the author need not request it and no time limit applies, because the closure is XSci's act. The reviewer is paid for work properly done before the referral, and nothing is recovered from them.

10. Appeals

10.1 An appeal against a certification decision

10.1.1 · Four grounds only.

(a) this Procedure was not followed; (b) a reviewer had an undisclosed interest that would have barred them under clause 3.3; (c) the report contains an error of fact about the manuscript; (d) the report does not meet the standard of detail this Procedure requires (clause 7).

10.1.2 · Disagreement with the reviewer's judgement is not a ground, and that includes the judgement on fit.

10.1.3 · How it is filed and decided. An appeal is filed with XSci, stating which ground or grounds are relied on and what in the report or the record supports them. There is no fee. It is decided by a person who took no part in the assessment, who may see the report, the record, the assignment and eligibility documentation, and any representations by the author; that person may not ask the reviewer to change the decision, and may not substitute their own view of the manuscript.

10.1.4 · The remedy. An appeal either succeeds or does not succeed. Where it succeeds, the author is refunded the assessment fee, and that is the whole of the remedy — no re-review, no replacement reviewer, and no change to the decision. The report stands on the record as written, and the decision does not change.

10.1.5 · A successful appeal also affects the reviewer's payment, because each of the four grounds is a finding that the assessment was not delivered as required: the payment for that assessment is not due, and is recovered if already paid, under the Reviewer Agreement. The reviewer is told the ground and may answer it before anything is withheld, without limit on the exchanges; where a question of fact cannot be resolved between the reviewer and the author, XSci adjudicates and may take an independent expert view. None of this turns on the decision the reviewer reached, and the author's refund does not depend on it — XSci owes the refund in full either way.

10.2 Where the successful ground is ground 2

10.2.1 · Two things follow together. Where an appeal succeeds on ground 10.1.1(b) and the version was certified: the certification is revoked (clause 11), because a record that does not disclose a barring interest is materially false; and the author is refunded.

10.2.2 · The refund is the whole of the remedy. No replacement reviewer is appointed and the version is not assessed a second time: XSci does not offer a re-assessment of an already-assessed version, and cannot process one. An author who wants a decision revises, deposits a new version, and buys an assessment of that version on the ordinary terms.

10.3 An appeal against a revocation

10.3.1 · A different set of standards applies. An appeal against a decision to revoke a certification does not lie on the four grounds in 10.1.1. It is decided against the guidance of the Committee on Publication Ethics (COPE), which governs in the first instance, together with the recommendations of the International Committee of Medical Journal Editors (ICMJE) and the applicable Crossref metadata standards, by a person who took no part in the revocation decision.

10.3.2 · No fee and no refund. The remedy is that the revocation is set aside and the certification restored to the record, with the notice corrected to show that it was set aside and the date; or that the revocation stands and the record is unchanged.

11. Revoking or correcting a record

11.1 · A certification is revoked only on a finding that its record was materially false — for example, that a reviewer held a barring interest and did not declare it, that the assessment recorded was not in fact carried out, or that the record misstates what was assessed. It is never revoked because a reviewer changed their mind, because later work or a failure to replicate disagreed, because the manuscript became controversial, or because a revision has since been deposited.

11.2 · A revoked certification is never deleted. The record remains publicly visible, marked as revoked, with the date and a reason.

11.3 · Errors are corrected; judgements are not. A factual error on a record is corrected with the original retained alongside it. A judgement is never corrected: a reviewer wishing to add or qualify something after delivery does so as a dated addition, and the original stands.

11.4 · Revocation does not affect the deposited work, which remains available at its DOI, dated as before.

12. The record, and what is published

12.1 · What the record holds. A completed assessment produces a permanent record: the version assessed, the date, the target venue, the version of the Standard and of this Procedure applied, the decision and its reasons, the findings and sentences for each domain, the verifiability step reached, the reviewer's competing-interest declaration, whether the review was blind, and the reviewer's own use of AI.

12.2 · A certification is always public, whether or not the report behind it is.

12.3 · Whether the report is published is the author's decision, on the terms in clause 2. Where an author declines to publish a refusal, it leaves no public trace at all.

12.4 · Nothing is ever published about an individual reviewer — not their turnaround, not their certification rate, not anything. Aggregate figures about the service are published at least annually and contain nothing per-reviewer.

12.5 · A reviewer takes no credit on the manuscript. Assessing is paid work and is never authorship, contributorship or acknowledgement.

13. Version of this Procedure

13.1 · Dated and versioned. A certification record names the version in force on the date the assessment was purchased, and a later change to this Procedure never reaches back into an assessment already bought. Superseded versions remain available so that an appeal can be measured against the text that applied.

13.2 · Related documents. The Certification Standard, the Peer Review Policy, the Reviewer Agreement, the AI and Data Usage Disclosure, the Retraction and Correction Policy, the Refund Policy and the Terms of Service.

13.3 · Contact: editorial@xsci.com