Science 'Under Review': What Each Status Means and Realistic Timelines

Science's system shows distinct status labels as your paper moves through each phase, which means "Under Review" actually tells you something specific: reviewers have your manuscript.

Science receives around 12,000 submissions per year and publishes roughly 800 research articles. Here's what editors filter for during those first 1-2 weeks:

• Broad significance. Will researchers outside your subfield care? Science isn't a specialty journal.
• Novelty, not just quality. A technically perfect study that confirms what people suspected doesn't clear the bar. Editors want results that change how people think about a problem.
• Complete evidence. Science papers are short (~4,500 words for Research Articles), but the data package needs to be airtight. Key claims resting on a single experiment without orthogonal validation are red flags.
• Clarity of the advance. Can the editor summarize what's new in one sentence?

A desk rejection from Science doesn't mean your work is bad. It means the editors didn't see it fitting the very specific niche that Science occupies: papers that will change how a broad scientific audience thinks about something. Plenty of excellent science gets desk-rejected here and lands well at field-specific top journals.

Science uses a Board of Reviewing Editors (BoRE), which is different from how Nature or Cell handle things. The BoRE includes around 150 active scientists who serve as the first line of evaluation after professional editors. When your paper arrives, a BoRE member with relevant expertise reads it and recommends whether to send it for full review. This means your desk decision isn't made by a professional editor alone, it's informed by a working scientist in a related field.

This system has a practical upside: the person evaluating your paper at the desk stage actually understands the technical content. The downside is that BoRE members are busy academics, which can occasionally add a few days to the desk decision timeline.

Reviewer selection. The handling editor (often the same BoRE member who recommended review) suggests 2-3 reviewers. Science typically uses 2-3 external reviewers per paper. They're chosen for direct expertise, and the journal makes an effort to avoid conflicts of interest. Reviewer invitations can take a week or more by themselves, since not everyone says yes on the first ask.

What reviewers evaluate:

• Are the central claims supported by the data?
• Are there alternative explanations the authors haven't ruled out?
• Is the experimental design appropriate for the claims?
• Does the significance match what Science publishes?
• Are methods reproducible and statistical analysis sound?

The short-format factor. At ~4,500 words, reviewers pay close attention to supplementary materials, that's where the full experimental detail lives. A disorganized supplement will cost you goodwill.

Review turnaround. Science asks reviewers to return reports within 10-14 days. In practice, it's common for at least one reviewer to take 3-4 weeks. The journal's editorial staff follows up actively, but they can't force anyone to finish on time.

Every Science reviewer form includes a section for "confidential comments to the editor." This is where reviewers say what they really think, without diplomatic softening. You won't see these comments, but they heavily influence the editor's decision. If your public reviews are lukewarm but you still get a revision request, the confidential comments were probably more positive than the formal report suggested.

Accept. Rare on the first round. Almost all Science acceptances go through at least one revision cycle. If you get a first-round accept, you've done something unusual.

Minor Revision. The editors and reviewers are satisfied with the core findings but want small changes: clarified language, an additional statistical test, a figure revision. You'll typically get 2-4 weeks to turn this around. Papers that receive minor revision requests are almost always accepted.

Major Revision. The most common positive outcome at Science. Reviewers see real value but want substantial additional evidence or analysis. Major revisions often include requests for new experiments, additional controls, or reanalysis of existing data. You'll usually get 2-3 months, sometimes more. A major revision request is genuinely good news, the editors wouldn't invest the time if they didn't believe the paper could reach the bar. That said, you need to address every point. Don't skip comments you find annoying or think are wrong. Respond to each one directly, even if your response is a respectful disagreement with evidence.

Reject After Review. Roughly 40-50% of reviewed papers are ultimately turned down. Common reasons include: reviewers identified a fundamental flaw in methodology or interpretation, the significance on closer inspection wasn't as broad as the desk screen suggested, competing work appeared during the review period, or reviewers couldn't agree and the editor sided with the more critical assessment. Rejection after review hurts more than a desk rejection because you've invested months, but it comes with detailed reviewer feedback from top experts. Those reports are often more thorough than what less selective journals provide. Use them, even if you disagree with the decision, the technical feedback will strengthen the paper wherever it lands next.

Reject with Encouragement to Resubmit. Science occasionally uses this category, and it's confusing because it sounds like a soft rejection. In practice, it means the editors think the core idea is strong enough for Science but the current evidence package doesn't get there. They're telling you: "Come back with more data." This isn't a polite brushoff. If you can address the major concerns (which usually means 6-12 months of additional work), a resubmission has a real shot. Treat the original reviewer comments as your revision roadmap.

Science and Nature are often lumped together, but their editorial cultures aren't identical. Science's BoRE system means desk decisions can be more technically informed, while Nature's professional editors may lean more on editorial judgment about audience appeal. In practice, Science desk decisions involve a working scientist making a formal recommendation, while Nature desk calls rely on editorial team judgment without a formal BoRE structure.

If you get revision requests:

1. Create a point-by-point response. Every reviewer comment gets a numbered response. Don't merge or paraphrase their points.
2. Do the experiments they ask for. "We feel this is beyond the scope" rarely goes well at Science.
3. Highlight changes clearly. Use tracked changes or colored text. Make it easy for reviewers to find what's different.
4. Respect the word limit. If you need to add material, you may need to cut something else or move it to the supplement.
5. Submit on time. If you can't meet the deadline, email the editor before it passes.

Science Advances deserves special mention. It's published by the same organization (AAAS), and there's a transfer option. When Science rejects a paper, the editor may suggest transferring to Science Advances. If you authorize the transfer, your manuscript, reviewer reports, and editor notes all move over. The Science Advances editors will evaluate the package and may accept without additional review, request a light revision, or send to one additional reviewer. Most transferred papers that address original reviewer concerns are accepted within 4-8 weeks. Don't treat it as a step down, Science Advances publishes strong multidisciplinary work and is open access, which often means more reads than behind Science's paywall.

Submissions sent in late November through early January often experience longer reviewer turnaround because of holiday schedules. June through August can also be slow as reviewers travel to conferences. If you have flexibility, early fall (September-October) and late winter (February-March) tend to produce the fastest turnaround. That said, don't delay a strong submission just to time the calendar.

• 0-3 weeks after submission: Don't contact the journal. Normal desk review time.
• Under Review for 4-6 weeks: Expected range. No need to follow up.
• Under Review for 8+ weeks: A polite status inquiry is reasonable. Keep it brief: "I'm writing to check on the status of manuscript SCI-XXXX."
• Decision Pending for 2+ weeks: Unusual. Worth a gentle check-in.

When you do follow up, email the editorial office rather than the handling editor directly. Be concise. Don't ask why it's taking so long, just ask for a status update. The delay is almost always caused by a reviewer who hasn't returned their report, and the editors are probably already chasing them.

• Science journal guide
• Science cover letter example
• Science acceptance rate
• Nature under review
• Cell under review
• PNAS under review
• Acceptance rates across journals
• Review timelines compared

Science submission readiness check. It takes about 1-2 minutes.

In our pre-submission review work with manuscripts targeting Science, three failure patterns generate the most consistent desk rejections. We find these across manuscripts we've reviewed through our Science submission readiness check.

The important finding framed too narrowly in the abstract. Science editors evaluate significance from the abstract before reading the paper. We observe that papers with findings that matter to a broad audience but presented in subfield-specific language generate desk rejections for insufficient breadth. The Science abstract must make the significance intelligible to a chemist reading a biology paper, or a physicist reading an ecology paper. Papers that open with the broadest formulation of the finding and use field-specific language only to support that framing clear the desk at substantially higher rates than papers that lead with the subfield context.

The conclusion whose scope exceeds the experimental basis. Science reviewers evaluate whether the evidence matches the scale of the claim. We find that papers claiming evolutionary significance from a single-species comparison, or mechanistic universality from experiments in one cell line, generate reviewer requests for expanded validation in roughly half of cases. SciRev community data for Science consistently shows "conclusion scope exceeds the experimental basis" as a recurring revision theme. Papers that either narrow the conclusion to match the experiments or include a second validation arm supporting the broader claim pass this standard consistently.

The missing control that any careful reviewer will identify. Science reviewers spend 4 to 8 hours reading a submission and are asked to evaluate whether the authors have ruled out alternative explanations. We observe that the single most predictable revision request at Science is a missing negative control, an unaddressed confound, or an alternative interpretation the experimental design does not rule out. Papers that include a concise "we considered and excluded alternative X by doing Y" section in the methods or discussion preempt this request. Papers that do not include explicit confound controls receive this as the first-listed major concern from reviewers in the majority of cases we review.