Most advice on how to write a grant proposal for a nonprofit hands you the same eight sections in the same order and calls it a day. The problem is that reviewers do not read your proposal the way you wrote it. They read it the way they score it, criterion by criterion, points in hand. Two organizations doing equally good work routinely land 12 points apart because one wrote to the rubric and the other wrote to its mission. This guide treats the rubric as the blueprint, then layers in the federal cost rules that quietly changed in 2024 and now decide what you can put in your budget.
The short version:
- Reviewers score against a published, weighted rubric. Allocate narrative space to point values, not to what you find most interesting.
- The 2024 Uniform Guidance revisions raised the de minimis indirect cost rate to 15%, the equipment threshold to $10,000, and the Single Audit trigger to $1,000,000 for budget years starting on or after October 1, 2024.
- Data collection and program evaluation are now explicitly allowable costs, so you can fund the outcome tracking that reviewers reward.
- Needs statements win on local data and a clean logic model, not national statistics or mission language.
- Match the ask, the population, and the geography to the funder before you write a word.
Why Reviewers Score Your Proposal Before They Read It
Federal programs publish a scoring matrix inside every Notice of Funding Opportunity, and most foundations work from an internal rubric even when they never show it to you. The score is not a formality applied after a careful read. It is the read. A reviewer typically holds a dozen or more applications, assigns points by criterion, and ranks the stack. The proposal that collects the most points wins, regardless of whose program a neutral observer would call more worthy.
That has a direct consequence for how to write a grant proposal for a nonprofit: every paragraph should be earning points against a named criterion, or it should be cut. The federal cost framework reinforces how literal this is. Under the 2024 revisions to the Uniform Guidance (2 CFR Part 200), the Environmental Protection Agency confirms the de minimis indirect cost rate rose to 15 percent and the Single Audit threshold to $1,000,000 for awards starting on or after October 1, 2024. Those are not stylistic preferences. They are the numbers a reviewer checks your budget against, and getting them wrong costs you in the section that is easiest to score objectively.
Read the Rubric First, Then Allocate Your Words
Before drafting, find the scoring criteria and the points attached to each. Federal NOFOs print them; for foundations, the closest proxies are the application questions and the program’s stated priorities. Then build your outline as a map of point values. If “project design” is worth 30 points and “organizational capacity” is worth 15, your project design section should be roughly twice as long and twice as detailed. Writing a beautiful, lengthy history of your founding while spending one thin paragraph on evaluation is how strong organizations leave points on the table.
This is also where funder fit gets decided. A reviewer cannot award points for alignment that is not there. Before committing time, confirm the ask size, the target population, and the geography match the funder’s history. Tools like the OpenGrants funder directory and a searchable grant database let you check a funder’s typical award range and priorities before you draft, so you are not requesting $400,000 from a program that caps grants at $50,000.
Where the Points Actually Live
Across federal programs at agencies like HHS, NIH, and DOE, a recurring pattern shows up in the criteria weighting. Industry analysis of published NOFOs puts the needs statement around 20 to 30 points and project design around 25 to 35 points, with evaluation, capacity, and budget splitting the remainder. The exact split varies by program, which is the point: read the actual rubric for the actual opportunity, then write proportionally. As one reviewer-focused breakdown puts it bluntly, the rubric is a gift because it tells you exactly what to write.
The 2024 Cost Rules That Change What You Can Budget
The budget section is where nonprofits most often lose easy points, and it changed materially in 2024. The Office of Management and Budget’s revisions to 2 CFR Part 200, summarized in OMB’s implementation guidance for federal agencies, took effect for awards on or after October 1, 2024. Three changes matter for almost every nonprofit budget.
First, the de minimis indirect cost rate is now 15 percent, up from 10. If you do not hold a negotiated indirect cost rate agreement, you may claim 15 percent of modified total direct costs without justifying it line by line. That is real money many organizations are still leaving uncollected out of habit. Second, the threshold that separates “equipment” from “supplies” rose to $10,000, which simplifies how mid-priced purchases are categorized and depreciated. Third, and most useful for scoring, costs related to gathering and managing data and to conducting evaluation are now explicitly allowable. You can budget for the outcome tracking that reviewers increasingly demand instead of absorbing it as unfunded overhead.
For federally funded work, treat these rules as part of the proposal craft, not an afterthought handed to your bookkeeper. The federal grants landscape rewards applicants who budget evaluation and indirect costs correctly, because a clean, compliant budget signals an organization that can actually manage the award.
Build the Needs Statement and Logic Model to the Score
The needs statement is where most proposals are won or lost, and the failure mode is predictable: applicants describe a national problem to justify a local intervention. Reviewers are not funding the national figure. They are funding a specific population in a specific place, so the data has to be local. Pull county-level numbers from the Census, your state health or education department, or local administrative records, and pair them with one concrete human detail. National statistics establish context; local data earns points.
From there, connect activities to outcomes with a logic model: inputs lead to activities, activities to outputs, outputs to outcomes. The distinction reviewers care about most is outputs versus outcomes. “We served 500 meals” is an output. “82 percent of participants reported stable food access at six months” is an outcome, and outcomes are what funders buy. Real funders now write this expectation into their guidelines. The Stranahan Foundation’s spring 2026 funding cycle, for instance, requires applicants to present a clear logic model and an evaluation plan spanning process, formative, and outcome evaluation. If you serve a defined community, your local nonprofit grants strategy should start from the outcome data you can already credibly report.
Organizational Capacity and Budget: The Sections Nonprofits Underweight
Capacity is the section applicants treat as a formality and reviewers treat as risk assessment. The reviewer is asking one question: if we send this money, will the work actually happen? Answer it with evidence, not adjectives. Name the staff who will run the program and their relevant track record, describe the financial systems that separate restricted from unrestricted funds, and reference past awards of similar size that you managed and closed out cleanly. A functioning board, fund-level accounting, and a data system that can produce the metrics you promised are worth more than any sentence about passion.
On the budget, every line must tie back to an activity in your methods section, and every number must be defensible. Reviewers read budgets for two signals: are the costs reasonable for the scope, and did the organization budget enough to do the work well. Under-budgeting reads as poor planning; over-budgeting reads as padding. Use the 15 percent de minimis rate openly if you have no negotiated rate, fund your evaluation explicitly now that it is allowable, and keep the math flawless. The budget is the one section a reviewer can check against external rules, which makes it the easiest place to lose points and the easiest place to keep them.
Frequently Asked Questions
How long should a nonprofit grant proposal be?
Follow the funder’s page limit exactly; exceeding it can disqualify you before scoring. Within that limit, length should track the rubric. If project design carries the most points, it gets the most space. Federal proposals often run 15 to 25 narrative pages, while foundation proposals may be three to five pages or a short online form. Never pad to fill space, and never compress a high-point section to make room for organizational backstory.
What is the most common reason nonprofit proposals get rejected?
Weak alignment and weak needs documentation. Many proposals lose before the narrative is read because the ask size, population, or geography does not match the funder. Among proposals that clear that bar, the frequent failures are using national data for a local problem, confusing outputs with outcomes, and budget errors. Each of these maps to a specific scored criterion, which is why writing to the rubric prevents most of them.
Do I need a logic model for every grant?
Not every application requires one, but building a logic model before you write is worth it even when it is not requested. It forces you to connect activities to measurable outcomes, which is exactly what evaluation and project-design criteria reward. Many 2026 funders, including foundations, now expect a logic model or theory of change, and developing one during the application crunch produces a rushed, unconvincing version. Build it during program design instead.
Can a nonprofit charge indirect costs on a federal grant?
Yes. If you have a negotiated indirect cost rate agreement, use it. If you do not, the 2024 Uniform Guidance lets you claim a 15 percent de minimis rate on modified total direct costs without a separate justification, up from the previous 10 percent. Many nonprofits still default to the old rate or skip indirect costs entirely, which leaves recoverable money on the table.
How early should I start writing?
Start the moment you confirm fit, because the substantive sections take time. Read the full NOFO or guidelines, build an outline mapped to the scoring criteria, and draft the needs statement and project design first. Save the executive summary for last, after every other section is final, so it accurately summarizes what you actually proposed.
Bottom Line and Next Steps
Knowing how to write a grant proposal for a nonprofit comes down to one decisive shift: stop writing to your mission and start writing to the score. Find the rubric, allocate your words to the point values, ground the needs statement in local data, and build a logic model that ends in outcomes a reviewer can verify. Then use the 2024 cost rules in your favor by claiming the 15 percent de minimis rate and budgeting the evaluation work that is now both allowable and rewarded.
If your team is strong on programs but stretched thin on proposal craft, that gap is exactly where points slip. OpenGrants’ managed grant writing services can help you translate solid programs into rubric-aligned, compliant proposals, so the score finally reflects the quality of the work. Pick your next opportunity, pull its scoring criteria, and build the outline around the points before you write the first sentence

