Grant Writing Coach

SKILL.md

---
name: grant-writing-coach
description: Coach non-profits, researchers, artists, and founders through writing grant applications, foundation proposals, and government funding requests. Diagnoses why proposals get rejected, rewrites the narrative, structures budgets, builds logic models / theory of change, and matches projects to funders. Knows the difference between US private foundations, federal grants (NIH/NSF/NEA), state/city grants, corporate giving, and international funders. Adapts advice for the four major proposal types: project, general operating, capital, and capacity-building. Use when asked to write a grant, find a funder, structure a budget, build a logic model, write a letter of inquiry, prepare a NSF/NIH application, address reviewer feedback, or rebuild a rejected proposal. Triggers on "grant writing", "grant proposal", "foundation grant", "nih grant", "nsf grant", "letter of inquiry", "letter of intent", "rfp response", "project proposal", "capacity grant", "capital campaign", "non-profit funding", "research funding".
metadata:
  tags: ["grants", "non-profit", "fundraising", "foundations", "research-funding", "writing", "proposals"]
---

# Grant Writing Coach

Coach a grant writer through the parts that actually move proposals from "thoughtful submission" to "funded." Built for founders of small non-profits, researchers writing their first NIH/NSF, program directors managing a grant pipeline, and individual artists hunting fellowships.

## Usage

**Basic invocation:**
> Help me find funders for [project / topic]
> Write a Letter of Inquiry for [foundation / project]
> Why are my proposals getting rejected?
> Build a logic model for my program
> Structure my grant budget

**With context:**
> Small non-profit, 3 staff, $400k annual budget, want to start grant pipeline.
> Postdoc applying for first R01, mentor's lab is well-funded but I need own grants.
> Visual artist applying for state arts council fellowship, $25k category.
> Mid-size org ($2M budget), 60% individual donations, 20% earned, 20% grants — want grants to 35%.
> Just got rejected by Robert Wood Johnson — debrief and apply to next round.

The coach starts by understanding the project, the funder ecosystem, and the writer's stage, then walks through the appropriate proposal type.

## Path Selection

Different funder ecosystems, different rules:

### Private foundations (US)

- **Examples:** Ford, Hewlett, MacArthur, Robert Wood Johnson, Open Society
- **Average award:** $25k–$2M
- **Process:** Often start with LOI; full proposal by invitation
- **Cycle:** Quarterly to annual, varies wildly
- **Best for:** mission-aligned projects, established or emerging non-profits, fellowship grants

### Federal grants (US)

- **Examples:** NIH, NSF, NEA, NEH, USDA, EPA, ED
- **Average award:** $25k–$5M+
- **Process:** RFP-driven; long-form proposal; rigorous review
- **Cycle:** 1–3 deadlines/year per program
- **Best for:** research, large programs, capacity-building (in some agencies)

### State and city grants

- **Examples:** state arts councils, city arts/cultural commissions, state DOH/DOJ
- **Average award:** $1k–$100k
- **Process:** Variable; often simpler than federal
- **Cycle:** Annual mostly
- **Best for:** local programs, emerging artists, small non-profits

### Corporate giving / corporate foundations

- **Examples:** Google.org, Salesforce.org, Microsoft Philanthropies
- **Average award:** $5k–$500k
- **Process:** Often tied to local presence or alignment with corp causes
- **Best for:** alignment with corp mission, matching gifts, employee programs

### Community foundations

- **Examples:** Silicon Valley Community Foundation, Boston Foundation
- **Average award:** $5k–$50k
- **Process:** Smaller ask, often simpler proposal
- **Best for:** local non-profits getting started, rapid-response

### International / multilateral

- **Examples:** Gates Foundation, Open Philanthropy, MacKenzie Scott (irregular), UN agencies
- **Average award:** Highly variable
- **Process:** Variable; often sole-source for established orgs
- **Best for:** global health, large-scale, established orgs

## Funder Matching

The first job: find funders who actually fund what you do. Most rejections come from misalignment, not weak proposals.

**Research tools:**

- **Candid (Foundation Directory Online):** $200–500/yr; comprehensive funder database
- **Instrumentl:** $179+/mo; modern UX, fit-scoring
- **GrantStation:** mid-tier; good for state/regional
- **Free:** Foundation 990 PDFs (search ProPublica Nonprofit Explorer); Grants.gov for federal
- **NIH Reporter / NSF Award Search:** see what's been funded recently

**Match criteria (must hit all to apply):**

- **Mission alignment:** funder's published priorities overlap with your project
- **Geographic fit:** funder funds in your region
- **Funding type:** project / general operating / capacity / capital — they fund yours
- **Award range:** their typical grant size matches your ask (don't ask Ford for $5k or a community foundation for $500k)
- **Active:** they've made grants in last 24 months
- **Eligibility:** 501(c)(3), fiscal sponsor accepted, individual eligible, geographic restriction

**Red flags in funder match:**

- Their last grant in your area was 5 years ago
- They've narrowed focus; your work isn't on the new priority list
- They only fund existing grantees and you're new
- Their "letters of interest accepted" page hasn't been updated in 18 months

## Letter of Inquiry (LOI)

Many private foundations require an LOI before a full proposal. 1–2 pages.

**LOI structure (1.5 pages):**

1. **Header:** organization, contact, funder name, request amount, project name
2. **Opening hook (1 paragraph):** why this issue matters, in 4–5 sentences
3. **Organization (1 paragraph):** mission, age, key accomplishments, why we're qualified
4. **Project (2 paragraphs):** what we'll do, who benefits, what changes
5. **Alignment (1 paragraph):** how this matches funder's priorities (specific reference to their published goals)
6. **Budget (1 sentence):** total project cost, what we're asking, what other support is in place
7. **Close:** invitation to discuss, contact

**LOI rules:**

- Reference funder's specific priorities by name
- Numbers, not adjectives ("served 1,200 youth in 2025" beats "many young people")
- Specific outcomes ("graduation rate increase from 60% to 75%")
- One outcome metric, not five
- Active voice
- No jargon

**LOIs that get full-proposal invitations:**

- Show clear cause-effect logic
- Demonstrate organizational capacity (staff, prior outcomes, board)
- Match the funder's voice and depth
- Don't oversell

## Full Proposal Anatomy

When invited or applying via RFP. Standard sections:

### 1. Executive summary (1 page)

The "tl;dr" — fundable on its own. Cover: who you are, what you'll do, why it matters, who benefits, total cost, ask amount, expected outcomes. Most reviewers read this and skim the rest. Make it strong.

### 2. Statement of need (2–3 pages)

Frame the problem with:

- Quantitative: data on the issue's scope (national + local)
- Qualitative: voices from the affected community
- Specific: not "youth need help" but "in [city/zip], 38% of high schoolers are below grade level in math, and dropout-from-college is 22%"
- Connected: why your project addresses this specific gap

Avoid:

- "There is no [solution] available" — usually false; reviewers know the field
- Long literature review (this is a proposal, not a paper)
- Grandiose framing ("world hunger" → narrow it down)

### 3. Project description (3–5 pages)

The heart of the proposal:

- **Goals (3–5):** outcome-level changes you'll create
- **Objectives (5–10):** measurable, time-bound milestones
- **Activities:** specific programs, services, events
- **Timeline:** Gantt or milestone chart
- **Staff:** who's doing what (briefly)
- **Beneficiaries:** specific population, numbers served
- **Theory of change:** the if-then logic (if we do X, then Y will change, because Z)

### 4. Logic model / theory of change (1 page graphic + 1 page narrative)

Standard 5-column structure:

```
Inputs → Activities → Outputs → Short-term outcomes → Long-term outcomes
$, staff, partners → workshops, events → # served → knowledge/skill change → behavior/condition change
```

For research grants: aims + hypotheses + experimental design replace this section.

### 5. Evaluation plan (1–2 pages)

- What you'll measure (metrics tied to outcomes)
- How you'll measure (surveys, observations, admin data)
- Timeline (baseline + check-ins + post)
- Who's responsible
- How findings will be used
- Independent vs internal evaluation

Strong evaluation plan = funder confidence. Weak = "they don't know if their work works."

### 6. Sustainability (1 page)

How this work continues after this grant. Funders fear funding a project that dies in 18 months.

- Diversified funding plan
- Earned revenue components
- Embedded into organizational core
- Replication / scaling potential

### 7. Organizational capacity (1–2 pages)

- Mission, history, key accomplishments
- Board governance
- Staff qualifications (key personnel only)
- Financial health (audit summary, ratios)
- Past similar grants successfully managed

### 8. Budget and budget narrative (2–3 pages)

Budget table + line-by-line narrative. Funders read this carefully.

**Budget categories:**

- Personnel (salary + fringe; show % of FTE)
- Consultants and contractors
- Direct project costs (materials, travel, technology)
- Evaluation
- Indirect / overhead (negotiated rate or de minimis 10%)
- In-kind contributions (clearly marked)

**Budget rules:**

- Ratios reasonable: personnel 60–75%, indirect <20%, evaluation 5–10%
- All federal grants: follow 2 CFR 200 cost principles
- Justify every line; reviewers question vague items
- Match (cash or in-kind) clearly identified
- Total cost vs ask: show full project cost AND what this grant funds

### 9. Attachments

- 501(c)(3) determination letter (US non-profits)
- Most recent audit or 990
- Board list with affiliations
- Key staff bios / CVs
- Letters of support (3–5; include partner orgs, beneficiaries, funders)
- Logic model
- Other (varies by funder)

## NIH / NSF Specifics

Federal research grants have unique rules:

### NIH (R01, R21, K-awards, etc.)

- **Specific Aims (1 page):** the most important page. 3 specific aims with hypothesis, rationale, expected outcomes.
- **Research Strategy (12 pages):** Significance + Innovation + Approach. Approach is the heaviest section — preliminary data, design, analysis, pitfalls, alternatives.
- **Biosketch:** 5-page max with personal statement, positions, contributions to science, scholastic performance.
- **Scoring:** 1–9 scale, lower better. Funded scores typically 10–30 percentile, varies by institute.
- **Resubmission:** A1 (one resubmission) only if reviewers ask for it; address every critique with explicit citation.

### NSF (varies by directorate)

- **Project Description (15 pages):** Intellectual Merit + Broader Impacts.
- **Broader Impacts:** non-trivial weight; have a real plan, not a checkbox.
- **Project Summary (1 page):** must explicitly address both criteria.
- **Data Management Plan, Postdoc Mentoring Plan, Results from Prior Support** — required attachments.

**Common federal mistakes:**

- Burying the hypothesis (it should be on page 1 of Specific Aims)
- Ignoring page limits (auto-rejection)
- No preliminary data (R01 reviewers expect it)
- Vague timeline / no contingencies
- Underestimating reviewer expertise (they know the field; don't oversimplify)

## Common Diagnoses

### "Rejected for misalignment"

- Funder priorities shifted; you applied to old guidance
- Your project is adjacent to but not within their scope
- Geographic mismatch you missed

Fix: read funder's recent grants (last 12 months) before applying; explicit alignment statement in cover letter.

### "Rejected for weak narrative"

- Problem statement too generic
- Logic model unclear
- Evaluation plan vague
- Budget unclear or too high

Fix: rewrite executive summary first; build clearer logic model; tighten evaluation; review budget for waste.

### "Capacity concerns"

- Budget too big for your org's track record
- Project scope larger than your demonstrated capacity
- Staff plan unclear

Fix: scale ask appropriately; partner with larger org for capacity grant; bring on advisor for credibility.

### "Got reviewer comments — need resubmission"

- Read every comment carefully; don't dismiss any
- Map each comment to specific revision in the resubmission
- Cite the comment ("Reviewer 2 noted X; we have addressed this by Y")
- Add "Response to Reviewers" section if format allows
- Don't fight; concede where they're right, defend politely where you disagree

## Calendar and Pipeline

Effective grant programs run on a calendar:

- **Annual:** plan top 10 funders for the year
- **Quarterly:** review pipeline, submit drafts
- **Monthly:** track new RFPs, deadlines
- **Weekly:** writing time blocked for active proposals

**Pipeline target (small non-profit):**

- 30–50 prospects researched/year
- 10–20 LOIs/year
- 5–10 full proposals/year
- 30–50% conversion to funded

Don't go for "spray and pray" — quality of fit beats volume.

## Output Format

The coach returns:

1. **Funder match assessment** — your project vs candidate funders
2. **Stage-appropriate plan** — LOI / full proposal / resubmission
3. **Outline** — section-by-section, with word counts
4. **Key narrative draft** — opening hook + executive summary
5. **Logic model framework** — inputs/activities/outputs/outcomes
6. **Budget framework** — major categories with rough %
7. **Reviewer-perspective check** — what they'll question
8. **Resubmission strategy** (if applicable) — how to address feedback