Non-ProfitFree
Why Your Grant Application Was Rejected
The lowest funded application scored 87, so the reviewer comments that look fixable cannot close the gap and rewriting is the wrong response.
Reviewer comments are prose and a decline is arithmetic, so River converts one into the other. In the worked example the reconstructed score is 68 of 100 against published weights. Of the 32 points lost, 17 are recoverable by writing better and 15 are not, because they rest on an evidence base and a track record that do not exist yet. A perfect rewrite therefore ceilings at 85, and the lowest funded application scored 87.
That two-point gap is the answer. It is not a judgement about how much better the narrative could be, it is the largest score this organisation could have submitted, and it loses. So the decision stops being whether to rewrite and becomes whether to change the ask, the partner or the funder. The same program funded 14 of 210 applications, at a mean award of $1.4M against a request here of $2.4M.
Written for the development lead deciding, six weeks out, whether the next cycle is worth the capacity. It reads the criteria the notice published, the selection factors it published separately, and the funder's last three years of actual awards. What comes back is a ceiling score and a reapplication call the board can follow. Before any resubmission goes out, the portal pre-check reads the package against both rule sets. The commercial version of this debrief is the RFP loss review, and the narrative behind it lives in a proposal content refresh.
The weights are published, and so are the reasons that have nothing to do with you
Federal notices are required to state how the criteria are weighted. The Uniform Guidance directs that where criteria vary in importance, an announcement should specify the relative percentages, weights, or other means used to distinguish among them. That makes a score reconstruction possible rather than speculative. Map every reviewer sentence to the criterion it addresses, weight it, and the vague sense that the proposal was weak becomes 68 out of 100 with the missing 32 points located and priced.
The same regulation separates three different reasons an application fails, and only one of them is about your writing. Eligibility is screened before review. Merit criteria are what reviewers judge. Then a notice must additionally list any program policy or other factors or elements, other than merit criteria, that the selecting official may use, giving geographical dispersion and program balance as its own examples. A strong application in a state that already has a grantee is a published outcome, not a mystery.
For a private foundation the fit test is arithmetic too, from a document you can already read. Form 990-PF requires the foundation to list all contributions, grants, etc., actually paid during the year with a specific stated purpose for each. Three years of that filing is a distribution. In the worked example it holds 118 grants to 71 organisations, only four at or above $150,000, and none of those four to a first-time grantee. The planned $250,000 ask sits above 96.6% of everything they have ever paid.
How it works
Send the decline
Whatever feedback arrived, including the versions that are two sentences long or internally contradictory.
Send the notice
The criteria, their weights, and the section listing selection factors other than merit.
Send the proposal
What you actually submitted, plus the award list if the funder published one.
Get the decision
A reconstructed score, a rewrite ceiling, and a stated recommendation with the arithmetic attached.
What you get
- Your score reconstructed criterion by criterion against the weights the notice published
- Every lost point split into what a rewrite recovers and what it cannot
- The ceiling a perfect rewrite reaches, next to the lowest application actually funded
- The decline tested against selection factors and oversubscription before it is blamed on prose
- Your ask sized against the funder's own grant distribution from its public filings
- An explicit reapply, reapply differently, or go elsewhere, with the arithmetic behind it
Common questions
The feedback is two sentences long. Is that enough to work with?
Usually, because the criteria and their weights are published even when the scores are not. Every reviewer sentence maps to a criterion, and criteria the reviewers said nothing about are themselves evidence. Thin feedback produces a wider band on the reconstructed score, which the document states rather than hides.
How can you know what the lowest funded application scored?
Often you cannot exactly, and where the funder publishes an award list, a payline or a range, that becomes the line. Where nothing is published the line is estimated from the award count against the application count and stated as an estimate. The reapply decision is then reported at a band rather than a point.
Why split lost points into recoverable and not?
Because they are different kinds of problem and only one responds to a deadline. A missing timeline is a writing gap. No prior award at this size is a fact about your history. Here 17 of the 32 lost points respond to a rewrite and 15 do not, which is what puts a ceiling on the resubmission.
Could we have been declined for reasons unrelated to the proposal?
Frequently, and federal notices are required to say so in advance. Alongside the merit criteria a notice must list the other factors a selecting official may use, and geographical dispersion and program balance are the regulation's own examples. Fourteen awards went to fourteen states here, and one was already taken.
How do you size an ask to a foundation we have never approached?
From their Form 990-PF, which lists every grant paid with its amount and stated purpose. Three years of filings is a distribution you can read. In this example 118 grants ran to 71 organisations, only four reached $150,000, and none of those four went to a first-time grantee.
What does the recommendation actually look like?
One of three, with the arithmetic attached: reapply as is, reapply with named changes to the ask or partners, or take it elsewhere. Here it is the third and the second together, because a $35,000 evaluation grant buys the comparison condition worth 6.5 points on the criterion that scored worst.
Is any of this useful before we get declined?
It is the same reading, done earlier and more cheaply. The criteria weights, the selection factors and the funder's own grant distribution are all published before the deadline. Running them against a draft is how you find out that a request is 171% of the program's mean award while there is still time to change it.
Why Your Grant Application Was Rejected
Fill in the form and your workspace opens with the work already underway.