Common grant writing mistakes are not all equal. A few quietly bleed points across every reviewer’s scoring sheet; others just annoy a reader and never change the final score. The difference matters because most writers fix mistakes in the order they notice them — polishing the easy stuff and leaving the score-killers in place. Treating “be more concise” and “outcomes are unmeasurable” as equivalent fixes is how strong programs lose grants they should have won.

This piece maps the most expensive common grant writing mistakes to the scoring sections reviewers actually use — NIH’s Simplified Peer Review Framework, NSF’s two merit-review criteria, and the rubric most foundations now publish. For applications due on or after January 25, 2025, NIH scores grant submissions under its Simplified Peer Review Framework, which reorganizes the old five criteria into three factors on a 1-to-9 scale, with 1 exceptional and 9 poor. Knowing where each mistake lands on that sheet is how you stop fixing the wrong things first.

Reviewers do not just notice mistakes — they score them

Every funded grant runs through a scoring instrument. At NIH, most research project grants are now reviewed under the Simplified Peer Review Framework announced in NOT-OD-24-010, which collapses Significance, Innovation, Approach, Investigator, and Environment into three factors: Importance of the Research, Rigor and Feasibility, and Expertise and Resources. Factors 1 and 2 get numerical 1-to-9 scores; Factor 3 is judged sufficient or not. Individual criterion scores are then folded into an overall impact score that runs from 10 to 90 after averaging and multiplying by ten.

At NSF, the Proposal & Award Policies and Procedures Guide requires reviewers to evaluate every proposal against two criteria — Intellectual Merit and Broader Impacts. Foundations vary, but most publish a rubric covering alignment, need, program design, organizational capacity, evaluation, and budget. Each line on each rubric is where a specific mistake lives or dies. A misplaced sentence in the needs statement does not register on the budget line and the budget line will not save it. The order in which you fix problems matters as much as the fixes themselves.

This piece walks through five common grant writing mistakes, the scoring line each one touches, what the reviewer comment usually sounds like, and the fix. For the compliance gauntlet that comes before scoring even starts, see our companion piece on why most grant applications get rejected early.

Mistake one: a problem or aims section that describes a topic, not a gap

The single most expensive mistake in the body of a proposal is a problem statement that describes an issue without documenting it, or — at NIH — a Specific Aims page that lists topics rather than testable gaps. At NIH, this falls under Factor 1 (Importance of the Research): reviewers ask whether the application addresses an important gap in knowledge, would solve a critical problem, or create a valuable conceptual or technical advance. At NSF, this is Intellectual Merit. At a foundation, this is the needs and alignment section — usually 20 to 30 percent of the total score.

Reviewer comments on this mistake are recognizable. They sound like “the proposal describes a topic rather than a specific gap” or “the needs statement relies on national figures and does not establish the local problem.” The Grantsmanship Center, which has trained grant professionals for more than fifty years, puts this at the top of its list of proposal failures: a problem statement that fails to document scope, ground itself in the served community, and connect logically to the proposed solution.

The fix is mechanical, not creative. Anchor every claim in data, use local statistics where possible, and make the scale of the problem match the scale of what you propose. For NIH applicants, the Specific Aims page should name the gap, the hypothesis, and the testable aims — and stop there. Reviewers form their first impression on that page, and it colors the rest of the read.

Mistake two: a budget that does not match the narrative

Budget-narrative mismatch is the most common form of credibility damage in a proposal. It also happens to be one of the cheapest mistakes to fix and one of the most expensive to leave in. At NIH, the budget shows up under Additional Review Considerations — reviewers may be asked whether the budget is sufficiently justified by the proposed project. At foundations, budget is usually a discrete rubric line worth 10 to 20 percent of the total, scored on whether costs are reasonable for the scope and whether the organization has budgeted enough to actually run the program.

Reviewer comments here are blunt. “The narrative describes hiring two case managers but the budget has one.” “A $250,000 budget does not match a project that clearly requires $500,000 of effort.” “Indirect costs appear hidden inside supply line items.” Every one of those is a direct hit on the budget line and an indirect hit on organizational capacity, because a budget that does not match the narrative reads as a sign the applicant did not actually plan the program. Reviewers cross-reference, sometimes line by line.

The fix is to build the budget and the narrative in the same week, not in sequence. Every staff line needs an explicit effort percentage — “Program Director, 25% FTE, $18,750” is acceptable; “Program Director, $18,750” is not, and federal grants routinely reject the latter. Every activity in the narrative needs a corresponding budget line, and every budget line needs a one-sentence justification with calculation logic. Use your federally negotiated indirect cost rate if you have one; otherwise, most funders accept a transparent 10 to 15 percent. For worked examples, see our walkthrough on writing a grant budget narrative.

Mistake three: outcomes that are actually outputs

This is the mistake that loses points everywhere. At NIH, it lands in Factor 2 (Rigor and Feasibility) — reviewers evaluate whether the proposed work is likely to produce compelling, reproducible findings within the proposed timeframes. At NSF, it lives inside the assessment of how well the proposed activities are well-reasoned and incorporate a mechanism to assess success. At a foundation, evaluation is usually 15 to 25 percent of the rubric on its own, and it is where many otherwise strong proposals fall apart.

The mistake is structural and easy to spot once you know what to look for. “We will serve 200 people” is an output. “75 percent of participants will demonstrate improved financial literacy by program end, measured by pre- and post-test scores on the standardized FFEL assessment” is an outcome. Reviewers write things like “outcomes are not measurable (no specific metrics, no timeframe, no baseline)” or “the evaluation plan relies on self-reporting without any verification mechanism.” Both of those are direct point deductions on the evaluation line.

What a credible evaluation plan looks like

A credible plan specifies, for each outcome: the metric, the instrument or data source, the person responsible for collecting it, the timepoints at which it will be collected, and how the data will be used to improve the program. If an external evaluator is involved, name them. Reviewers reward methodological rigor, and they recognize evaluation sections written as a learning tool rather than a compliance afterthought. Most importantly, every outcome should map to a goal in the program description, and every goal should map to a budget line that funds the activity producing the outcome.

Mistake four: writing to your own mission instead of the funder’s criteria

The most strategic of the common grant writing mistakes is also the hardest to see in your own draft. It is the habit of writing the proposal in the organization’s internal language, around the organization’s internal priorities, and leaving the reviewer to translate. Funders fund solutions to problems they care about — not organizational sustainability, not staff credentials in the abstract. A proposal structured around the applicant’s identity rather than the funder’s published criteria starts every reviewer’s scoring at a deficit.

This mistake touches alignment scoring at foundations (often the single largest rubric line, 25 to 30 percent) and the Significance section at NIH. Reviewer comments sound like “the technical work does not genuinely address the funder’s stated priorities” or “the proposal reads like a square peg forced into a round hole.” Once a reviewer notices this, skepticism colors the rest of the read and depresses scores in sections that would otherwise have scored well.

The fix is to read the funder’s review criteria before drafting and again before submitting, and to allocate narrative space proportionally to the points each section carries. Use the funder’s section names as H2 headings when the application allows free formatting. For foundation work, read the most recent annual report and giving list, not just the mission page. Our directory of specialized grant writers includes practitioners who do rubric-alignment work for a living.

Mistake five: skipping a structural requirement that triggers return without review

The most preventable common grant writing mistake is the one that prevents your proposal from being scored at all. NSF maintains a published list of reasons it will return a proposal without review: missing Broader Impacts as a separate section of the Project Description, missing Results from Prior NSF Support, exceeding page limits, non-compliant formatting, or duplicating a proposal already under review. None of those are scoring issues. They are gates that close before scoring begins.

Reviewer comments do not apply because the proposal never reaches a reviewer, and there is no appeals process for an NSF return without review. NIH, foundations, and state grantmakers have their own version: missing attachments, wrong file formats, page limits enforced by stopping reading at the limit, signature pages absent. The proposal sits outside the contest entirely.

The fix is a compliance checklist built from the funder’s instructions before you write the first sentence, run against the package at least two days before the deadline by someone other than the author. For federal applicants, confirm SAM.gov registration is active, the Unique Entity ID is current, and the Grants.gov workspace is set up. The gate before the contest is its own discipline.

How to score your own draft before reviewers do

A pre-submission audit takes about three hours and changes the score more than another revision pass. Pull the funder’s published review criteria into a single document, list each scoring section with its point weight, then read your draft as if you were the reviewer, scoring each section on the funder’s scale. Be ungenerous. The goal is to find the sections that would score below midrange and revise those, not to feel good about the sections already scoring well.

Then count narrative pages by section and compare the count to the point allocation. If significance is worth 30 points but fills two pages, while budget at 10 points fills four, the proposal is misallocating reviewer attention. Trim the long sections, expand the short ones, and run the compliance checklist one final time.

Frequently Asked Questions

Which of the common grant writing mistakes costs the most points?

Across federal and foundation reviews, the most expensive single mistake is writing to your own mission rather than the funder’s published criteria, because it depresses scores in every section that depends on alignment — and that is most sections. The second most expensive is a problem statement or aims section that describes a topic without naming a gap, because it sets up a low first impression that colors the rest of the review. Both are fixable in a single revision pass.

Do foundation reviewers actually use a written rubric?

Most professional foundations now do, and many publish it. Even smaller community foundations typically use a shared rubric for staff reviews. If the rubric is not published on the funder’s website, the most recent NOFO or guidelines will usually describe the review criteria in enough detail to back into the scoring weights. When in doubt, ask the program officer what reviewers are scored against — most are willing to tell you in a five-minute call before the deadline.

How do I tell whether an outcome is really an output?

Ask whether the statement describes a change in the participant or community. “Served 200 people” describes a unit of activity. “Reduced housing instability for 200 households by 30 percent over the program year, measured by the Arnold housing-stability index” describes a change you can measure. Outputs belong in the activity section; outcomes belong in the evaluation section. Conflating them is the most common evaluation-section error reviewers flag.

What does an NIH 1-to-9 score actually mean for my chances?

On NIH’s scale, 1 is exceptional and 9 is poor. The final overall impact score is the mean of eligible panelists’ scores multiplied by 10, producing a value between 10 and 90. Funding lines vary by institute and fiscal year, but in most institutes only the strongest scoring quartile is funded outright. Applications judged in the bottom half are typically “not discussed” and receive no overall impact score at all.

If my proposal was rejected, should I resubmit or move on?

That depends on the reviewer comments. Critiques that cluster around clarity, detail, justification, or organization point to writing and presentation problems, which a revision can fix. Critiques that cluster around significance, innovation, impact, or alignment suggest the idea or the funder choice needs to change. Federal program officers are usually willing to advise on whether a specific application is competitive for a resubmission.

Bottom line: score your draft before reviewers do

The common grant writing mistakes that cost the most points are not the ones that sound the worst when read aloud. They are the ones that pull scores down in the sections worth the most points: alignment, problem statement or specific aims, evaluation, and budget. Fixing them in order — alignment first, then problem framing, then evaluation, then budget — recovers more score than any amount of line editing.

The best signal that a proposal is ready to submit is that the writer has read it once with the funder’s rubric in hand and revised every section that would currently score below midrange. That single discipline separates funded proposals from the ones that lose narrowly. For teams that want a second pair of eyes on a draft before submission, our grant writing services include rubric-based pre-submission review, with a focus on the scoring sections where most common grant writing mistakes actually live.