How Long Does Peer Review Take? Why Time to Decision Is the Wrong Number

Peer review usually takes about three months to a first decision, and the range runs from days to more than half a year depending on the field and the journal. In a study of 3,500 author-reported review experiences, the average first response took 13 weeks (Huisman & Smits, 2017): 8–9 weeks in medicine and public health, 11 in the natural sciences, and 16–18 in the social sciences, humanities, mathematics and computer science, and economics and business. Journals that publish their own timings give the view from the inside. PLOS ONE reported a median of 45 days to first decision and 188 days to acceptance for January to June 2023 (PLOS ONE journal information).

Authors usually stop there. Editors and publishers have a harder job: deciding whether a shorter number means better review, or only a faster process. The rest of this article sets out the timeline first and then spends its time on that second question, with practitioners who spoke on Prophy Roundtable, the monthly panel of editors and publishing leaders convened by Prophy, which builds reviewer-finding and conflict-of-interest tools for publishers.

How long does peer review take, field by field?

The most useful public dataset on author-side duration is the Huisman and Smits analysis of 3,500 reviews that researchers reported to the SciRev platform. It is self-reported and self-selected, so it describes experience rather than a census, but it is large and it separates fields.

• Medicine and public health: 8–9 weeks to a first response.
• Natural sciences and general journals: 11 weeks.
• Psychology: 14 weeks.
• Social sciences, humanities, mathematics and computer science, economics and business: 16–18 weeks.

The spread inside a field is wide too. In social sciences and humanities only 7–8% of authors heard back within a month, against 25% in the natural sciences and 27–28% in medicine and public health. In 10% of all reported cases authors waited more than six months. Rejected manuscripts took about four weeks longer to get a first response than accepted ones.

Time to first decision is also not time to publication. PLOS ONE's 2023 medians run from 17 days to a first editorial decision (rejection or sending the paper to review), through 87 days to a final decision, to 188 days to acceptance. Anyone comparing journals should check that both numbers measure the same stage.

Where do the weeks go?

A first decision bundles several waits: editorial triage, the search for reviewers, the review itself, and the editor's decision. Only one of them belongs to reviewers.

Huisman and Smits found that one-third of journals take more than two weeks to deliver an immediate desk rejection, and one in six takes more than four. Their reading: besides the time reviewers take, inefficient editorial processes play an important role. A desk decision involves no reviewer at all.

The reviewer search is the other wait, and it restarts every time an invitation is declined or ignored. Two diagnoses of the reviewer problem came out of Prophy Roundtable's July episode, and they never fully merged. One treats it as a mismatch between supply and demand: submissions compound while reviewer time stays flat. The other treats it as reachability. Lucia Steele, Publications Manager at AboutScience, put the second view this way: "the reviewers out there receive a lot of invitations, … and obviously they have to be very selective in their replies" (#6, 22:52). Reviewers exist, and the right one is hard to reach in time.

The size of this wait shows up in invitation figures. In the Biology Open pilot covered below, invitations to a pre-contracted, paid reviewer pool were accepted 67% of the time, against 23% for conventional review requests (Research Information). Whatever the cause, an editorial office that tracks invitations sent per accepted review can see this wait and start shortening it. Matching the manuscript to people with the right expertise and no conflict of interest sits at the front of that search, and it is where Prophy works. For editorial departments that need to find high-quality peer reviewers quickly, Prophy offers an extensive database of over 104 million researchers' profiles; Referee Finder is the tool built on it.

Does a shorter time to decision mean better review?

It measures speed and nothing else. Tony Alves, Co-Founder of Scholarly Publishing Solutions, spoke on Prophy Roundtable about what the metric records: "it tells us how quickly a journal completed a transaction, not whether the review, the peer review, improved the science" (#8, 16:36).

For an editor that has a practical consequence. A board report that opens with median time to decision rewards the workflow and says nothing about the manuscript. Report what changed between submission and acceptance next to the clock: how many reviewer comments led to new analyses, how many revisions altered a conclusion. The clock stays in the report. It stops being the headline.

Which dates should a journal disclose?

All three: submission, acceptance and publication. Kent Anderson, Founder & CEO of Caldera Publishing Solutions, argued on the same panel that cross-field comparisons mislead because fields run on different clocks: "I don't think there is a speed of science. I think science happens" (#8, 9:23). His proposal was routine disclosure, so that each community can read its own record. Some platforms do not show those dates, and without them nobody can later study how a community behaves.

The SciRev field spread above is the argument in numbers. A mathematics journal at 16–18 weeks is not slower than a medical journal at 8–9 in any sense that matters to its authors, who expect their field's pace. Huisman and Smits found exactly that: authors in the fields with the longest processes rated them highest. Disclosure lets an editor compare the journal against its own history and its own community, not against an industry average that mixes botany with orthopaedics.

What can a speed gain be credited to?

Rarely to one lever. The much-cited case is a paid-review model piloted at Biology Open from 2024. The journal reported that, for peer-reviewed manuscripts submitted in 2025, average time to a first decision based on completed reviews fell from 37.7 working days under conventional review to 5.5 under its Fast & Fair model, a cut of about 85% (Research Information).

Selina La Barbera, Publishing Director, Journals at EDP Sciences, cited the result on Prophy Roundtable's episode on paid peer review and attached the caveat herself. The experiment also changed reviewer recruitment, deadlines and quality checks, "so we cannot attribute the improvement to payment alone" (#9, 13:29).

For an editor planning a pilot of any kind, the lesson is procedural: change one thing at a time, and measure report quality and cost next to turnaround. A package of four changes can show a large speed gain and leave open which of the four earned it, or whether quality moved at all.

Which steps should be faster, and which should be slower?

Some of both. Joy Moore, Principal at Joy Moore Consulting Ltd, described the choice editors face when tools arrive: add technical "lubricants" to speed work through, and then ask "how do you fine-tune them, and how do you maybe introduce a little bit more friction?" (#8, 47:17). Her point was that the answer differs journal by journal and portfolio by portfolio.

A workable rule for an editorial office: pick one step this month and decide on purpose whether it should be faster or slower. A faster reviewer search costs authors nothing, and a faster desk decision helps them when it is accurate: a quick "not for this journal" lets an author move on without losing weeks (more in our editor's guide to desk rejection). Rushing the scientific assessment is where speed starts to cost something. Peer review at a breaking point goes further into that trade-off.

FAQ

How long does peer review take for a journal article?

About 13 weeks to a first response on average in the largest author-reported study, ranging from 8–9 weeks in medicine and public health to 16–18 weeks in the social sciences, humanities, mathematics and computer science, and economics and business. Time to acceptance is longer: PLOS ONE reported a median of 188 days for early 2023.

What counts as a good time to first decision?

A journal that publishes an expected window and meets it. Short medians help authors plan, but a median shared across a whole portfolio hides field differences, so the useful comparison is a journal against its own history and its own community.

Why does peer review take so long?

Triage, the reviewer search, the reviews and the editor's decision each take time. Author reports point to editorial handling as well as reviewers: one-third of journals take over two weeks to deliver a desk rejection. On the reviewer side, invitations that are declined or unanswered restart the search.

Does faster peer review mean better peer review?

Not by itself. A faster first decision shows a manuscript moved through a workflow. Whether review improved the paper has to be measured separately, as in changes between submission and acceptance.

Prophy helps publishers identify relevant reviewers, detect conflicts of interest, and discover related research using a database of 104M+ researcher profiles and 200M+ publication records. Learn more at prophy.ai.

About the experts

All five spoke on Prophy Roundtable, Prophy's monthly panel of editors and publishing leaders.

• Tony Alves — Co-Founder, Scholarly Publishing Solutions. Prophy Roundtable #8: The Speed–Quality Trade.
• Kent Anderson — Founder & CEO, Caldera Publishing Solutions. Prophy Roundtable #8: The Speed–Quality Trade.
• Joy Moore — Principal, Joy Moore Consulting Ltd. Prophy Roundtable #8: The Speed–Quality Trade.
• Selina La Barbera — Publishing Director, Journals, EDP Sciences. Prophy Roundtable #9: Should Peer Review Be Paid, Restructured, or Abolished?
• Lucia Steele — Publications Manager, AboutScience. Prophy Roundtable #6: The Reviewer Crisis.