For GRFP applicants before they submit
NSF GRFP Mock Review of Your Application
Reviewer-style comments on your GRFP personal statement and research plan, with both criteria rated, your own lines quoted and five ranked fixes.
Free to start. No card needed.
How to review a GRFP application yourself
Start with the skim. Each reviewer reads about 20 to 25 applications, according to NSF's instructions for reviewers1, so make your main points easy to find. List each heading, each paragraph's first sentence and any bold text, and read only that list. It should give your question, your hypothesis, what you have done and what you will do for others. Then circle any term a reader outside your subfield wouldn't know: NSF's tips for applicants2 say reviewers will generally know your field, but probably not your topic.
Next, read for evidence. NSF 26-5263 lists the abilities reviewers weigh for Intellectual Merit, from planning and conducting research to persisting through challenges. Beside each claim of skill, mark the sentence that proves it: a method you ran, a result with a number, a talk or a paper. Circle every "we" and ask whether a reader can tell your part from your lab's. In the plan, find the hypothesis, aims, methods and expected results, plus what you'd do if the main approach failed.
Test Broader Impacts the same way. Past activities need your role and numbers: how many students, for how long, what changed. Future plans need a who, a where and a how often, and should grow from what you've already done. Then check that both statements tell one story, and that each uses the exact headings "Intellectual Merit" and "Broader Impacts". Ratings sheets that past applicants shared show reviewers flagging an unclear personal role, a hypothesis never stated outright and Broader Impacts that were "not very concrete."
How it works
How do you get reviewer comments on a GRFP draft?
Drop your statements in the box above. River reads both in full, then writes reviewer-style comments into a doc in your workspace. For Intellectual Merit and Broader Impacts, it gives an estimated rating and lists strengths and weaknesses, each quoting your line and naming a factor NSF lists. A skim test, a check that both statements tell one story, a rules check and five ranked fixes follow. It never edits your files. It also can't confirm fonts or margins, so check your PDFs with the format checker.
For the 2026 competition, NSF's additional FAQ4 says reviewed applicants got a band rank and ratings for both criteria, but "No written reviewer comments were collected." So line up written feedback yourself before you submit, from a mentor using the GRFP review rubric, a writing center or a review like this one. Make sure it covers both criteria. NSF 26-5263 says each "is necessary but neither, by itself, is sufficient," so strong lab work cannot carry thin Broader Impacts.
The 2026–27 deadlines have passed. Applications closed October 19 to 23, 2026, and NSF will notify applicants by early April 2027. Applying next fall? Check NSF's next solicitation before relying on any rule here (this cycle's was posted August 25, 2026), and run your review a few weeks before your deadline. If Broader Impacts comes back weak, the Broader Impacts builder drafts it from what you've actually done. For a thin plan, the research proposal writer drafts a new one from your idea or a paper.
How it works
From draft to reviewer's notes
Drop your statements
Drop your personal statement and research plan as PDF or Word files, or paste them. One works too.
River reads both
It reads every paragraph first, working out your stage, your field and what you propose to study.
Get your review
A memo lands in your doc: both criteria with estimated ratings, the skim test, rule checks and five fixes.
Fix and recheck
Make the changes you agree with, or ask River to redraft your weakest paragraph in a new doc.
Research Plan · Intellectual Merit
We developed a pipeline to classify coral bleaching from drone images, and the results were promising.
In graduate school I will extend this work to new reef sites.
- Whose work?"We developed" hides your part. Say what you built.
- Promising how?Give a result a reviewer can weigh: a number or a finding.
What you'll get
A reviewer's read of both statements before you submit
- Intellectual Merit and Broader Impacts read like a reviewer, with your lines quoted
- A skim test: what a busy reviewer takes from your first sentences alone
- Your five most important fixes, in order, across both statements
Before and after
Your statements in. A reviewer's notes back.
A three-page personal statement and a two-page research plan as PDFs. A chemistry senior proposes a cheaper catalyst for breaking down plastic waste, after two summers of lab research and three years running a tutoring program.
Intellectual Merit · River's estimate: Very Good Strength: "I presented the results as a poster at a regional meeting" shows you can communicate research. Weakness: "we optimized the reaction" never says which step was yours. Broader Impacts · River's estimate: Good Strength: "I founded a weekly tutoring program that now serves 40 students" is concrete. Weakness: "I will do outreach in local schools" names no partner, schedule or numbers. Skim test: the hypothesis first appears on page two of the plan. Fix first: move the hypothesis into the plan's opening paragraph.
Ratings are River's estimate from your text, not a prediction of NSF's decision. Each strength and weakness quotes your words.
What lands in your doc
What's in your GRFP review
- Intellectual Merit strengths and weaknesses, each naming an NSF factor and quoting the line behind it
- Broader Impacts read the same way, with past activities and future plans checked for specifics
- River's estimated rating for each criterion, from Excellent to Poor, with one plain sentence on why
- A skim test of your headings and paragraph openings, and what a skimming reviewer would miss
- Where your two statements disagree, and where the plan ignores the experience you describe
- Your five most important fixes, ranked, starting with anything that could get your application returned without review
Questions GRFP applicants ask
Common questions
What do GRFP reviewers look for?
Evidence for NSF's two criteria. Intellectual Merit is your potential to advance knowledge, shown by abilities such as planning research, working independently and communicating results. Broader Impacts is your potential to benefit society, through activities such as mentoring and outreach. NSF tells reviewers1 that applications "are not research proposals, but fellowship applications," so they judge you on your whole record, not only the project.
Do GRFP applicants get reviewer comments?
Not for the 2026 competition (applications submitted in fall 2025). NSF gave reviewed applicants a band rank and ratings for both criteria, and collected no written comments. Ratings sheets from earlier competitions, quoted in the GRFP examples guide, included each reviewer's written explanation. NSF's pages for the 2026–27 cycle don't say whether written comments will return. The ratings decoder explains the band and ratings you got.
Is River's rating the same as NSF's?
No. River's rating is its own estimate from your text, on a five-step scale from Excellent to Poor, meant to show which criterion needs the most work. NSF's panels also read your letters, transcripts and the rest of your record, and rank you against other applicants in your field. Treat River's rating as a pointer to your weakest section, never as a forecast.
Is it OK to have AI review my GRFP statements?
Yes, as long as what you submit is yours. NSF 26-526 has no rule about AI tools, but it makes you certify that both statements are your own original work. River's review quotes your lines and suggests changes, and never rewrites your files. NSF's 2023 notice on generative AI5 adds that proposers answer for the accuracy and authenticity of anything AI helped produce.
How do I get a GRFP mock review with River?
Use the box at the top of this page: drop your personal statement and research plan as PDF or Word files, or paste the text. One statement works, though River then can't check that the two tell one story. River reads everything, then writes your review into a doc in your free River workspace, where you can ask it to redraft your weakest paragraph.
Is the GRFP mock review free?
Yes, it is free to start, and no card is needed. Your review opens in a River workspace, where you can edit it, share it or ask River to change it. Paid plans add more AI use and access to the most advanced models.
Are my statements used to train AI?
No. River's privacy policy says it does not use your documents or drafts to train AI models. The AI providers that process them work under agreements that bar training on your content, too. Your statements and the review stay in your River workspace, and you can delete them whenever you like.
Your next step