NSF does not use a single “annual report” form. Every NSF grant carries up to three distinct, separately-deadlined reports, all submitted through Research.gov: the annual project report, the final annual project report, and the Project Outcomes Report for the General Public (POR) — the only one of the three that NSF posts publicly, verbatim, as submitted. Missing any of them has real consequences: NSF will not process a no-cost extension, a funding increment, or a pending proposal from an identified PI/co-PI while a report on that award is overdue.
This guide covers what each report is, exactly when it is due, what it must and must not contain, and what happens if it’s late. All deadlines and requirements below are drawn directly from the NSF Proposal & Award Policies & Procedures Guide (PAPPG), Chapter VII.D, the current NSF 24-1 edition (with its 2025–2026 supplements).
The three NSF project reports at a glance
| Report | Deadline | Who sees it | Purpose |
|---|---|---|---|
| Annual project report | No later than 90 days before the end of the current budget period | NSF program officer (internal) | Progress in the current budget period only — not cumulative |
| Final annual project report | No later than 120 days after the award end date | NSF program officer (internal, though summarized data may be released under FOIA) | Progress in the project’s final period; signals the PI considers the scope of work complete |
| Project Outcomes Report (POR) | No later than 120 days after the award end date | The general public — posted verbatim on Research.gov | Plain-language, 2–3 paragraph summary of intellectual merit and broader impacts for a lay reader |
Last verified 2026-08-16 against the NSF PAPPG (NSF 24-1), Chapter VII.D, “Reporting Requirements” — new.nsf.gov/policies/pappg/24-1/ch-7-award-administration. NSF updates the PAPPG periodically via numbered supplements; re-check the current PAPPG edition before treating exact day-counts as unchanged if you’re reading this more than a year after the verification date above.
Annual project reports: due 90 days before the budget period ends
NSF requires a project report for essentially every assistance award, submitted in Research.gov. The annual project report covers progress in the current budget period only — it is explicitly not cumulative, so a report for year two of a project should not restate year one’s accomplishments.
The submission window is easy to miscalculate because the PAPPG anchors it to the budget period, not the calendar year: the annual report must be submitted no later than 90 days before the end of the current budget period, to give the cognizant NSF program officer time to review and approve it before the period actually ends. If it isn’t submitted and approved by the time the budget period ends, the report is overdue.
A few structural details that change how many reports you’ll actually file:
- Multi-year standard grants require an annual report for each budget period as described above.
- Continuing grants are subject to the same reporting policy, with one wrinkle: if an increment runs 18 months or more, NSF requires two annual reports for that increment — one covering the first 12 months, and a second covering the remaining months. An increment of exactly 24 months requires one annual report per 12-month period within it.
- Cooperative agreements treat the annual project report as a gate: NSF will not approve a future funding increment until it’s submitted and approved.
- Fellowship awards may not require an annual project report at all — check the specific program solicitation, since fellowship reporting requirements are set at the solicitation level rather than by the general PAPPG rule.
Consequence of missing the window: NSF will delay processing of additional funding and administrative actions tied to the award — including no-cost extensions and, for continuing grants, the next funding increment — until the report is submitted and approved.
The final annual project report: due 120 days after the award ends
The final annual project report is the last report in the series and, like the annual reports before it, covers only the most recently completed budget period (still not cumulative). Unless the award specifies otherwise, it is due in Research.gov no later than 120 days after the award’s end date; the report is considered “due” throughout that 120-day window and becomes overdue only once the window closes without submission and approval.
Submitting the final annual project report has a specific meaning: the PI is signifying that the project’s scope of work is complete and that no further research activity — including a no-cost extension, supplemental funding, or an award transfer — is anticipated. It does not, however, block the recipient from requesting further payment for costs already incurred during the period of performance.
Within that same 120-day window, the recipient must also submit to the program officer any award-specific end items (reports or deliverables named in the award, solicitation, or an NSF guide/brochure it references) and, where the award carries mandatory cost sharing, a final cost-share notification certified by the Authorized Organizational Representative (AOR).
The stakes for lateness are higher here than for a routine annual report. Failure to submit final project reports on time delays NSF’s review and processing of any pending proposal for every identified PI and co-PI on that award — not just future actions on the award that’s overdue.
The Project Outcomes Report for the General Public (POR)
The POR is structurally different from the other two reports: it’s the only one written for a lay audience and the only one NSF makes public. It is due on the same 120-day-after-award-end-date schedule as the final annual project report, also submitted through Research.gov, and it becomes overdue the day after that 120-day window closes if the PI hasn’t submitted it.
What it must contain: a brief summary — generally two to three paragraphs — describing the project’s outcomes or findings, addressing both intellectual merit and broader impacts as defined in NSF’s merit review criteria, written in plain language. NSF explicitly encourages PIs to avoid jargon, terms of art, and acronyms. NSF automatically pulls in the publications already reported in the annual/final reports; PIs can also list other outputs such as data sets, software, collections, or educational materials.
What it must not contain:
- Confidential or proprietary business information
- Unpublished conclusions or data that could compromise the ability to publish in the research literature later
- Invention disclosures that could adversely affect patent rights (the PI’s or the institution’s)
- Personally identifiable information — home contact details, individual demographic data, or anything individually identifiable collected from human research participants
Once submitted, NSF posts the POR electronically exactly as written, with a standing disclaimer that the report reflects the PI’s own views and that NSF “has not approved or endorsed its content.” Anyone can look up an award’s POR through NSF’s public award search — this is the one NSF report a journalist, a member of Congress, or a curious member of the public is actually likely to read.
What every NSF project report is made of
NSF’s annual and final annual project reports follow the same structural components as the government-wide Research Performance Progress Report (RPPR) format also used by NIH and other federal research agencies:
- Accomplishments — what was done and what was learned in the reporting period
- Products — publications and other outputs generated
- Participants/Organizations — who worked on the project and any partner organizations
- Impact — how the work advanced the field, other disciplines, workforce development, or society
- Changes/Problems — deviations from the approved plan
- Special Requirements — only where the specific award or solicitation calls for them
This information feeds directly into NSF’s own accountability obligations: NSF uses it to compile the performance reporting Congress requires under the Government Performance and Results Act (GPRA) of 1993, and it can be released to the public through a Freedom of Information Act (FOIA) request even for the two reports that aren’t proactively published like the POR.
Certification: what submitting a report actually asserts
Every annual and final annual project report submission is a certification by the PI or co-PI that the statements in it (apart from scientific hypotheses and opinions) are true and complete, and that the report’s text and graphics — along with any accompanying publications or other documentation — do not misrepresent the project’s results. This is not a formality: it’s the same certification framework that underpins NSF’s research-integrity enforcement.
Overdue reports: what actually happens
| Overdue report | Practical consequence |
|---|---|
| Annual project report | No-cost extensions and other administrative actions on the award are delayed; for continuing grants, the next funding increment is delayed; for cooperative agreements, future increments are blocked outright |
| Final annual project report | NSF delays review and processing of any pending proposal naming an identified PI or co-PI on the overdue award — a consequence that reaches beyond the award itself |
| Project Outcomes Report | Becomes overdue the day after the 120-day window closes; a persistently overdue POR is visible in Research.gov’s own report-status tracking used by program officers |
The practical takeaway for a sponsored-programs office: track the annual report’s 90-day-before-period-end trigger separately from the two 120-day-after-award-end deadlines, since the annual report clock runs on a completely different reference point (the budget period) than the final report and POR clocks (the award end date).
Where reports are submitted
All three reports are submitted through Research.gov, which has been NSF’s system of record for project reporting since March 2013 — well before FastLane was fully retired as a proposal-submission option in January 2023. There is no separate paper or email reporting path; a report that isn’t in Research.gov, submitted by the PI and approved by the cognizant program officer, isn’t considered filed.
Frequently asked questions
Is the NSF annual project report the same as an RPPR?
Functionally yes. NSF’s project reports are designed to be fully consistent with the government-wide Research Performance Progress Report (RPPR) standard, using the same component structure (Accomplishments, Products, Participants/Organizations, Impact, Changes/Problems). NIH’s version of the same standard is covered in CASRAI’s NIH RPPR guide; NSF just calls its version the “annual project report” rather than “RPPR” in the PAPPG text, and it’s filed in Research.gov rather than eRA Commons.
Do I need to submit a Project Outcomes Report if my final annual project report already covers the results?
Yes — they’re separate submissions with separate purposes. The final annual project report is a detailed, internal progress report for the program officer; the POR is a short, public, plain-language summary. Both share the same 120-day-after-award-end deadline, but submitting one does not satisfy the requirement for the other.
Can I revise a Project Outcomes Report after it’s posted?
The PAPPG does not describe a public-facing revision process in the reporting-requirements section covered here; NSF posts the POR “exactly as it is submitted.” If your award requires a correction, raise it with your cognizant NSF program officer rather than assuming self-service editing is available.
What happens if I never submit a required NSF project report?
Beyond the delayed-funding and delayed-proposal-review consequences described above, an award with a chronically overdue final report is a closeout blocker — NSF cannot close out an award while required reports remain outstanding, and unresolved reporting is one of the standard items NSF and institutional sponsored-programs offices check before signing off on an award’s closeout.
Does the annual project report need to be cumulative across the whole project?
No. Both the annual project report and the final annual project report are explicitly written for the current (or most recently completed) budget period only. Cumulative project narrative belongs in the POR’s summary and in any renewal or continuation proposal, not in the annual reports themselves.
