How Funders Decide Who Gets Money

Funders decide grant awards in three stages: a legal screen for what the money may pay for at all, a merit review that scores technical quality, and a selection step where staff weigh scores against portfolio, policy, and risk. Technical merit qualifies an application. It rarely decides it.

Current figures — verified 2026-08-11

Item	Value	Source
NIH research project grant success rate	13.0% (FY2025)	NIH Extramural Nexus
NSF competitive-proposal funding rate, FY2025 (all competitively reviewed proposals)	19% — 8,377 awards from 43,532 proposals evaluated	NSF FY2025 Agency Financial Report

These figures change. Verify against the linked source before relying on them. Report an outdated figure

Key takeaways

• A grant award decision is a portfolio decision, not a ranking exercise.
• Statute and appropriation set the outer boundary of what is fundable.
• Peer review separates sound from flawed more reliably than good from excellent.
• Program policy factors and risk review sit between the score and the award.
• Foundations decide on a board calendar, with staff acting as advocates.

What limits a funder’s grant award decisions?

A grant award decision begins inside a boundary the funder did not set. A federal agency may award only what an authorizing statute permits and an appropriation funds, and it must use a grant agreement rather than a contract when the principal purpose is “transferring a thing of value” to carry out a public purpose (31 U.S.C. § 6304).

That boundary is invisible in most application guidance and decisive in practice. A project can be excellent, well evidenced, and cleanly budgeted, and still be unfundable because the authorizing language does not reach it. Reading the authorizing statute and the assistance listing before writing is the cheapest disqualification check available, and it is the one most applicants skip. How that legal layer sits inside the wider system is covered in Grant Funding Fundamentals.

Private funders operate under a different set of limits on the grant award decision. A private foundation making a grant to an individual for travel or study incurs a taxable expenditure unless the grant is awarded on an objective and nondiscriminatory basis under a procedure approved in advance by the IRS (IRS, Grants to Individuals). Grants to entities that are not public charities require expenditure responsibility, a documented process with its own reporting. Neither rule forbids the grant. Both make it expensive enough that most foundations decline by default, which is why a “we don’t fund that” answer is often a paperwork answer rather than a judgment about the work.

How does federal merit review score grant applications?

Federal merit review is a structured evaluation by subject-matter experts against published criteria. Agencies are required to run one: “the Federal agency must design and execute a merit review process of applications for discretionary Federal awards,” whose objective is “to select recipients most likely to be successful in delivering results” (2 CFR 200.205).

Two implementations dominate, and the difference matters for where influence sits. At the National Science Foundation, a proposal is reviewed “usually by three to ten other persons outside NSF either as ad hoc reviewers, panelists, or both,” after which the Program Officer writes a review analysis and “recommends to the cognizant NSF Division Director whether the proposal should be declined or recommended for award” (NSF PAPPG Chapter III). Program staff sit close to the decision.

At the National Institutes of Health, merit review is deliberately firewalled from program staff. A study section run by a Scientific Review Officer produces criterion scores and an overall impact score on a nine-point scale, where the final score is the mean of the voting members’ post-discussion scores multiplied by ten (NIH, Scoring System and Procedure). Applications in the lower half are not discussed and receive no overall impact score. A second tier, the Advisory Council, then weighs those scores against institute priorities before the institute director decides (NIH, Second Level of Review).

How reliable are grant peer review scores?

Grant peer review scores carry real signal and real noise, and honest planning requires holding both. The largest controlled replication of the NIH process found no measurable agreement among reviewers rating the same applications: “We found no agreement among reviewers in evaluating the same application,” across 43 reviewers, 25 applications, and 83 ratings (Pier et al., PNAS, 2018).

Read that finding precisely, because it is routinely overstated. The study drew only from high-quality applications, so it measures the inability to separate good from excellent — not an inability to separate sound work from flawed work. Reviewers do the second job reasonably well. The first job is where the noise lives, and the first job is exactly what a competitive funding line asks them to do.

Panel-level simulation shows the same effect at the award boundary. Reanalyzing 2,705 proposals scored by 45 panels in one Australian national funding round, researchers found that only “9% (n = 255) of grant proposals were always funded, 61% (n = 1662) never funded, and 29% (n=788) sometimes funded” once ordinary variation in panel composition was modeled (Graves, Barnett & Clarke, BMJ, 2011). Scores also predict downstream productivity weakly: across 102,740 funded NIH grants, percentile rank explained roughly one percent of the variance in later publication and citation outcomes (Fang, Bowen & Casadevall, eLife, 2016).

The practical consequence is a planning rule, not a reason for cynicism. A near-miss score is weak evidence that the work was weak. Treat a single decline as one draw from a noisy process, and treat repeated declines with consistent written criticism as signal.

Who makes the final grant award decision?

The final grant award decision belongs to an agency official, not to the review panel. Uniform Guidance requires each notice of funding opportunity to disclose “any program policy, factors, or elements that the selecting official may use in selecting applications for the award” and gives the standard examples: “geographical dispersion, program balance, or diversity” (2 CFR 200, Appendix I).

The National Science Foundation states the consequence plainly:

“A proposal doesn’t need to receive all ‘excellent’ scores to be funded. Conversely, receiving all ‘excellent’ scores is no guarantee of funding.” — NSF, Overview of the NSF Proposal and Award Process

A second gate sits after selection. Before making an award, an agency must evaluate the applicant’s risk, including “the applicant’s record of effectively managing financial risks, assets, and resources,” the quality of its management systems, its history on previous federal awards, and audit findings (2 CFR 200.206). An organization can win on merit and lose on a late audit, a lapsed registration, or an unresolved finding. Building internal controls for grant recipients is therefore part of competing, not only part of complying.

Portfolio fit is knowable in advance. Funded-award databases show what an agency has already bought, and a pre-application conversation with the program officer establishes whether a proposed project duplicates the portfolio or fills a gap in it. Applicants who read the Notice of Funding Opportunity for the selecting official’s factors, not only the scoring criteria, are reading the document the way the decision is actually made.

How do private foundations decide grant awards?

Private foundations decide grant awards through staff advocacy on a board calendar. A program officer builds a docket of recommended grants, writes the internal case for each, and presents it to trustees who meet on a fixed schedule — often quarterly, sometimes annually. Trustee preference, founder intent, and existing relationships shape which proposals reach the docket at all.

Two things follow for applicants. The program officer is an advocate, not a decider, so the job is to supply the officer with a defensible internal case: a named problem, evidence of capacity, a realistic ask, and one acknowledged risk with a mitigation. Proposals that project frictionless competence are harder to defend in a room than proposals that name a real constraint.

The second is that stated philosophy runs ahead of practice. The Center for Effective Philanthropy’s study of multiyear general operating support found “a sobering disconnect between attitudes of foundation leaders and the experience of nonprofits, as well as a similar disconnect between the attitudes of foundation CEOs themselves and their foundations’ practices” (CEP, New Attitudes, Old Practices). Verify a foundation’s behavior against its filed grant tables rather than its published values statement, a method covered in foundation grants.

What do applicants misread about grant decisions?

Applicants misread grant decisions in five recurring ways, and each has a specific cost.

• Treating the score as the decision. A score qualifies an application for the eligible set. Program policy factors, portfolio balance, and risk review determine what leaves that set as an award.
• Skipping portfolio research. Prior awards are public. An applicant who has not read what the program already funds is arguing in the dark about the one factor staff weigh most.
• Ignoring the risk record. A lapsed registration or an open audit finding can defeat a strong application after review, with no route back through the merit process.
• Appealing on the merits. Most agencies offer no appeal of scientific or technical judgment. Where reconsideration exists, it tests whether review was procedurally fair, not whether reviewers were right.
• Over-fitting to a single decline. Given documented panel variability, one decline near the funding line is weak evidence about the work. A pattern of consistent written criticism across reviews is strong evidence.

The correction that changes outcomes most is the first one. Applications built to score well and nothing else compete against a decision rule that was never only about the score. Guidance on interpreting reviewer comments and deciding whether to try again is in what to do when a grant is declined.

Frequently asked questions

Does the highest-scoring proposal always get the grant?

No. Federal notices must disclose the program policy factors a selecting official may apply after merit review, including geographic dispersion and program balance (2 CFR 200, Appendix I). The National Science Foundation states that all-excellent scores are “no guarantee of funding.”

Who actually reads a grant application at a federal agency?

External subject-matter reviewers read and score it, program staff read it and write the recommendation, and business or grants-management staff review the applicant’s financial and compliance record. At the National Science Foundation, a proposal is typically reviewed by three to ten people outside the agency.

Can you appeal a grant funding decision?

Rarely on the merits. Most federal agencies provide no appeal of technical judgment. Where a reconsideration process exists, it examines whether the review was fair and reasonable procedurally and substantively — not whether reviewers reached the correct scientific conclusion.

Why do two reviewers score the same proposal differently?

Because criteria are applied subjectively and reviewers weigh strengths and weaknesses differently. A full replication of the NIH review process found no measurable agreement among reviewers evaluating identical applications (Pier et al., PNAS, 2018).

Does contacting a program officer help a grant application?

It helps with fit and timing, which are the questions a notice cannot answer. Program officers can say whether a project belongs in their program, what the portfolio already contains, and whether a budget scale is plausible. They cannot disclose reviewer identities or individual reviewer comments.

Do foundations score proposals the way federal panels do?

Usually not. Foundation decisions run through staff recommendation and board approval on a meeting calendar rather than through a scored peer panel. Judgment, fit with trustee priorities, and the program officer’s internal advocacy carry most of the weight.

Related topics

• Grant Funding Fundamentals — the hub for how grant funding works
• What Is a Grant?
• The Grant Funding Lifecycle
• Grant Funding Myths, Corrected
• Writing for the Reviewer

Sources

1. U.S. Code. 31 U.S.C. § 6304 — Using grant agreements. Legal Information Institute, Cornell Law School. https://www.law.cornell.edu/uscode/text/31/6304 (accessed 2026-08-11)
2. Office of Management and Budget. 2 CFR 200.205 — Federal agency review of merit of proposals. eCFR. https://www.ecfr.gov/current/title-2/subtitle-A/chapter-II/part-200/subpart-C/section-200.205 (accessed 2026-08-11)
3. Office of Management and Budget. 2 CFR 200.206 — Federal agency review of risk posed by applicants. eCFR. https://www.ecfr.gov/current/title-2/subtitle-A/chapter-II/part-200/subpart-C/section-200.206 (accessed 2026-08-11)
4. Office of Management and Budget. 2 CFR Part 200, Appendix I — Notice of Funding Opportunity. eCFR. https://www.ecfr.gov/current/title-2/subtitle-A/chapter-II/part-200/appendix-Appendix%20I%20to%20Part%20200 (accessed 2026-08-11)
5. National Science Foundation. Proposal and Award Policies and Procedures Guide (PAPPG 24-1), Chapter III: NSF Proposal Processing and Review. https://www.nsf.gov/policies/pappg/24-1/ch-3-proposal-processing-review (accessed 2026-08-11)
6. National Science Foundation. Overview of the NSF Proposal and Award Process. https://www.nsf.gov/funding/overview (accessed 2026-08-11)
7. National Institutes of Health. Scoring System and Procedure. https://grants.nih.gov/grants/peer/guidelines_general/scoring_system_and_procedure.pdf (accessed 2026-08-11)
8. National Institutes of Health. Second Level of Review: Advisory Council or Board. https://grants.nih.gov/node/46 (accessed 2026-08-11)
9. Pier, E. L., Brauer, M., Filut, A., Kaatz, A., Raclaw, J., Nathan, M. J., Ford, C. E., & Carnes, M. Low agreement among reviewers evaluating the same NIH grant applications. Proceedings of the National Academy of Sciences 115(12): 2952–2957, 2018. https://www.pnas.org/doi/10.1073/pnas.1714379115 (accessed 2026-08-11)
10. Graves, N., Barnett, A. G., & Clarke, P. Funding grant proposals for scientific research: retrospective analysis of scores by members of grant review panel. BMJ 343:d4797, 2011. https://pubmed.ncbi.nlm.nih.gov/21951756/ (accessed 2026-08-11)
11. Fang, F. C., Bowen, A., & Casadevall, A. NIH peer review percentile scores are poorly predictive of grant productivity. eLife 5:e13323, 2016. https://elifesciences.org/articles/13323 (accessed 2026-08-11)
12. Internal Revenue Service. Grants to Individuals — Private Foundations. https://www.irs.gov/charities-non-profits/private-foundations/grants-to-individuals (accessed 2026-08-11)
13. National Institutes of Health, Office of Extramural Research. Fiscal Year 2025 by the Numbers: Extramural Grant Investments in Research. NIH Extramural Nexus, March 2026. https://grants.nih.gov/news-events/nih-extramural-nexus-news/2026/03/fiscal-year-2025-by-the-numbers-extramural-grant-investments-in-research (accessed 2026-08-11)
14. National Science Foundation. FY 2025 Agency Financial Report (competitive proposal, award, and funding-rate trends, Table 1.1). https://nsf-gov-resources.nsf.gov/files/FY-2025-Agency-Financial-Report_0.pdf (accessed 2026-08-11)
15. Center for Effective Philanthropy. New Attitudes, Old Practices: The Provision of Multiyear General Operating Support. 2020. https://cep.org/report/new-attitudes-old-practices/ (accessed 2026-08-11)