Your science is fine. Your margins aren't.
NIH may withdraw an application from consideration before a single reviewer reads it — for a font size, a page count, a header, a password. Grant Review Pro checks every attachment against NIH's actual format rules in seconds, and cites the rule it checked against.
— HOW GOOD APPLICATIONS DIE
The minimum font size. PDF conversion software routinely shrinks text below it without telling you. NIH checks. You don't.
The minimum margin, on all four sides. No applicant text may appear in them. A single figure caption bleeding into the margin is a formatting violation.
Reviewers who will read a withdrawn application.
Twelve months of work. Five to seven thousand dollars if you brought in a writer. A registration process that starts a year out. And the failure that ends it can be a footer.
— THE CHECKLIST
- ·Encrypted or password-protected PDFs
- ·Paper larger than 8.5 × 11
- ·Content inside the half-inch margins
- ·Headers and footers
- ·Page counts over the limit for your activity code
- ·Text under 11pt outside figures
- ·Type density over 15 characters per linear inch
- ·Line spacing over six lines per inch
- ·Scanned pages
- ·Revision markup in a resubmission
We fail only what we can prove from the file. Everything else is a warning for you to check. A tool that cries wolf three days before a deadline is worse than no tool.
— WHY GENERIC CHECKERS GET THIS WRONG
“SBIR Phase I” is not a page limit. R43 is — and its Research Strategy limit is six pages, while R44, SBIR Phase II, gets twelve. STTR splits the same way: R41 six, R42 twelve. The Commercialization Plan is a twelve-page requirement for Phase II and isn't expected at Phase I at all.
We key every rule to the activity code, then apply the precedence NIH actually uses: a Guide Notice overrides the NOFO, and the NOFO overrides the general application guide. When no limit exists in any of the three, there is no limit — and we don't invent one.
— TAKE IT TO YOUR RESEARCH OFFICE
A finding that says “page limit exceeded” is a linter. A finding that says “Research Strategy is 7 pages. R43 limit is 6. Source: NIH Table of Page Limits, verified 2026-06-22” is something you can act on, forward, and defend.
Every result shows the page, the excerpt, and the rule. We tell you when we last verified the rules against NIH, because NIH changes them.
Research Strategy is 7 pages. R43 limit is 6 pages.
— HONESTLY
This does not check your science.
It checks whether your science will be read. Run a Mock Study Section for the science.
This does not write or edit.
Reviewers recognize AI-written prose. We score, review, diagnose, and check. We never write.
A pass is not a guarantee.
We measure what a PDF can tell us. Your research office and the NOFO remain the final word.
— YOUR APPLICATION IS PRE-PUBLICATION IP
Unsubmitted Specific Aims are among the most confidential documents a company owns. We treat them that way.
- ›Your files are not used for training — ours or anyone's.
- ›No model sees them during this check. The compliance engine is a PDF parser, not an AI.
- ›Your PDF is uploaded to a server function, held in memory for the length of the check, and discarded — never written to disk or any storage bucket. Only the structural findings are saved.
- ›Encryption at rest is provided by our managed storage backend (Supabase). Findings are hard-deleted after 30 days unless you save them to your Vault, and on request at any time.
Find out before NIH does.
Upload one attachment. Sixty seconds. No card.
RULES VERIFIED AGAINST NIH FORMAT ATTACHMENTS AND TABLE OF PAGE LIMITS · LAST VERIFIED 2026-06-22 · SUBMISSION MATCH RESULTS AND COMPLIANCE CHECKS ARE NOT GUARANTEES OF FUNDING OR ACCEPTANCE.