Build an AI Grant Writing Service with Make.com: The Complete Step-by-Step Guide

Manual grant writing is for consultants who want to stay small forever. If you want to run an AI grant writing operation that generates $10,000+/month in recurring revenue, you need an automated engine that finds grant opportunities, drafts proposals, checks compliance, and tracks submissions — all without you spending 60 hours per proposal. This guide is the technical implementation manual. Follow every step in order. Skip nothing. If you skip the compliance checker, a formatting error could disqualify an otherwise winning proposal. If you skip the funder alignment analysis, you will write proposals that are technically correct but strategically misaligned with the funder’s priorities. Every step exists because someone lost a grant by skipping it.

This guide assumes you have zero infrastructure set up. By the end, you will have a fully automated pipeline that can handle 10+ clients and 30+ proposals per month with minimal manual intervention.

Prerequisites

Before you start, you need the following:

• Foundation Directory Online — $50/mo (funder database access)
• Make .com — $16/mo (Teams plan for 10,000 operations/month)
• OpenAI API key — $10 credit minimum (platform.openai.com/api-keys)
• Anthropic Claude API key — $10 credit minimum (console.anthropic.com)
• Google Workspace — $6/mo for professional email and Docs access
• Notion — Free (client dashboards and document management)
• Grammarly Business — $15/mo (grammar and style checking)
• 6-8 hours of uninterrupted time for initial setup

Total upfront cost: $97/mo + $20 API credit. A single client at $1,500/month covers this 13x over.

Step 1: Set Up Your Grant Research Pipeline

The foundation of your service is finding the right grants for each client. This module builds the automated grant research system that scans databases, matches opportunities to client profiles, and delivers prioritized recommendations.

Sub-step 1a: Create the Client Profile System

In Google Sheets, create a template called “Client Profile Template” with these sections:

Organization Information:

• Organization name, legal status (501c3, LLC, etc.), EIN
• Mission statement (full text)
• Programs and services (list with descriptions)
• Target population served
• Geographic service area
• Annual budget and fiscal year
• Key staff qualifications and credentials
• Partnerships and collaborations
• Previous grant history (funder, amount, year, outcome)

Grant Preferences:

• Minimum grant amount willing to pursue
• Maximum grant amount capable of managing
• Preferred grant types (program, general operating, capital, research)
• Geographic restrictions (local, state, regional, national, international)
• Timeline preferences (immediate funding vs. future planning)
• Willingness to apply for federal grants (requires SAM.gov registration)
• Budget for grant writing services

Save this template in Notion. Duplicate it for each new client and fill in the details during onboarding.

Sub-step 1b: Build the Grant Opportunity Scanner

Create a Make.com scenario: “Grant Scanner — [Client Name]”

1. Trigger: Schedule — Runs every Monday at 6 AM
2. Module 1 — HTTP Request: Query the Grants.gov API for new opportunities matching the client’s keywords and eligibility

Grants.gov API endpoint: https://api.grants.gov/v1/api/search2

Parameters:

keyword: [client's program keywords]
eligibility: [client's organization type]
status: "posted"
oppStatus: "forecasted|posted"
rows: 50

3. Module 2 — Code by Zapier : Parse the API response and extract: grant title, agency, description, award range, deadline, eligibility requirements, and CFDA number

4. Module 3 — OpenAI “Create a Chat Completion”:

System prompt:

You are a grant research specialist. Given a list of grant opportunities and a client's profile, score each opportunity on the following criteria:

1. Eligibility match (0-100): Does the client meet the eligibility requirements?
2. Program alignment (0-100): Does the grant fund the type of work the client does?
3. Competitiveness estimate (0-100): How competitive is this grant likely to be? (Lower = more competitive = harder to win)
4. Funder alignment (0-100): Based on the funder's history and stated priorities, how well does this match?
5. Timeline feasibility (0-100): Can a quality proposal be submitted by the deadline?

Calculate an overall match score as a weighted average:
Eligibility (30%) + Program (25%) + Funder (25%) + Timeline (10%) + Competitiveness (10%)

Respond in JSON:
{
  "opportunities": [
    {
      "title": "",
      "agency": "",
      "deadline": "",
      "award_range": "",
      "match_score": 0,
      "eligibility_match": 0,
      "program_alignment": 0,
      "competitiveness": 0,
      "funder_alignment": 0,
      "timeline_feasibility": 0,
      "recommendation": "APPLY|CONSIDER|SKIP",
      "reasoning": "2-3 sentence explanation"
    }
  ]
}

5. Module 4 — Notion: Add opportunities scoring above 60 to the client’s “Grant Pipeline” database
6. Module 5 — Gmail: Send weekly email with the top 5 new opportunities ranked by match score

Sub-step 1c: Supplement with Foundation Directory Research

Grants.gov covers federal grants. For foundation and corporate grants, manually search Foundation Directory Online weekly. Build a parallel workflow:

1. Search FDO for: funders in the client’s program area, geographic region, and organization type
2. Extract funder profiles: giving history, average grant size, application process, deadlines
3. Feed funder profiles through the same AI scoring system
4. Add high-scoring opportunities to the client’s Grant Pipeline

Step 1 Check-In

Verify each of these before moving on:

1. Client Profile Template created with all required fields
2. Grant Scanner scenario runs weekly and finds relevant opportunities
3. AI scoring produces accurate match scores (test with 10 known grants)
4. Top opportunities are added to the client’s Grant Pipeline in Notion
5. Weekly email delivers prioritized recommendations to the client
6. Foundation Directory research supplements federal grant search

Step 2: Build the AI Proposal Drafting Engine

This is the core of your service. The proposal drafting engine takes a grant opportunity and a client profile, and produces a comprehensive first draft that you refine and polish. The AI handles 80% of the writing; you handle the 20% that requires strategic judgment, organizational voice, and funder-specific positioning.

Sub-step 2a: Create the Proposal Template Library

In Notion, create a template library organized by funder type:

Federal Grant Template:

• Cover Page (SF-424)
• Executive Summary (1 page)
• Needs Assessment (3-5 pages)
• Program Design and Implementation (5-8 pages)
• Organizational Capacity (2-3 pages)
• Evaluation Plan (2-3 pages)
• Budget and Budget Narrative (3-5 pages)
• Letters of Support
• Required Attachments

Foundation Grant Template:

• Letter of Inquiry (2-3 pages)
• Full Proposal (8-15 pages)
• Budget Summary
• Organizational Background

Corporate Giving Template:

• Sponsorship Proposal (3-5 pages)
• Partnership Framework
• Impact Metrics

Each template includes placeholder text, section-specific writing guidelines, and common evaluation criteria for that funder type.

Sub-step 2b: Build the Drafting Workflow

Create a Make.com scenario: “Proposal Drafter — [Client Name]”

1. Trigger: Client approves a grant opportunity from the Grant Pipeline (status changes to “Approved for Application” in Notion)

2. Module 1 — HTTP Request: Download the full grant announcement (NOFO/RFP) from the funder’s website

3. Module 2 — OpenAI “Create a Chat Completion” (Requirements Analysis):

System prompt:

You are a grants compliance analyst. Given a grant announcement, extract and organize:

1. Eligibility requirements (who can apply)
2. Required proposal components (what must be included)
3. Page limits and formatting requirements
4. Evaluation criteria with weightings (how proposals are scored)
5. Submission method and deadline
6. Required attachments and forms
7. Key dates (webinar, Q&A deadline, submission deadline)
8. Special requirements (match funding, partnerships, registrations)

Respond in JSON format with each section clearly labeled.

4. Module 3 — Claude (Needs Assessment Draft):

Use Claude for the needs assessment — it excels at long-form, data-driven narrative.

System prompt:

You are an expert grant writer specializing in needs assessments. Given a client's organization profile and a grant opportunity, draft a compelling needs assessment section (3-5 pages).

Include:
- Opening paragraph connecting the funder's mission to the community need
- Current state data with specific statistics (cite sources)
- Gap analysis: what is missing in the current landscape
- How the client's programs address this gap
- Why now: urgency factors that make this the right time to act

Rules:
- Use specific numbers and data points, not vague claims
- Cite at least 3 authoritative data sources
- Connect every data point to the client's programs
- Write in the client's voice (professional, passionate, evidence-based)
- Never use "we believe" or "we feel" — use "data shows" and "evidence indicates"

5. Module 4 — OpenAI (Program Description Draft):

System prompt:

You are a grant writer drafting the program description section. Given the client's organization profile, the grant requirements, and the needs assessment, write a comprehensive program description (5-8 pages).

Include:
- Program overview and objectives (aligned with evaluation criteria)
- Implementation plan with timeline and milestones
- Staffing plan with roles and qualifications
- Partnership plan (if applicable)
- Sustainability plan (how the program continues after grant funding ends)

Rules:
- Every objective must be SMART (Specific, Measurable, Achievable, Relevant, Time-bound)
- Align each objective with the funder's evaluation criteria
- Include specific deliverables for each quarter
- Reference the client's track record of successful program implementation

6. Module 5 — OpenAI (Evaluation Plan Draft):

Draft an evaluation plan (2-3 pages) that includes:
- Logic model connecting activities to outputs to outcomes
- Measurable indicators for each objective
- Data collection methods and timeline
- Reporting schedule aligned with funder requirements
- Continuous improvement process

7. Module 6 — OpenAI (Budget Narrative Draft):

Draft a budget narrative (3-5 pages) that:
- Justifies every line item in the budget
- Connects each expense to a specific program activity
- Explains cost calculations (e.g., "Personnel: Program Director at $65,000/year × 50% effort = $32,500")
- Includes indirect cost rate explanation
- Aligns with the funder's budget requirements and allowable costs

8. Module 7 — Notion: Create a proposal draft page with all sections
9. Module 8 — Gmail: Notify you that the draft is ready for human review

Sub-step 2c: Build the Funder Alignment Analysis

Before you start writing, analyze the funder’s giving history to align your proposal with their actual priorities (not just their stated priorities).

Create a separate scenario: “Funder Analysis”

1. Trigger: New grant opportunity approved for application
2. Module 1 — OpenAI: Analyze the funder’s recent giving patterns

Given this funder's recent grant awards (from Form 990 and press releases), analyze:
1. Average grant size awarded
2. Types of organizations funded (by budget size, geographic focus, program area)
3. Geographic distribution of awards
4. Common themes across funded proposals
5. What differentiates successful applicants from unsuccessful ones

Provide 5 specific recommendations for positioning this proposal to align with the funder's demonstrated preferences.

3. Module 2 — Notion: Add the analysis to the proposal draft page

This analysis takes 2 minutes with AI and can increase win probability by 30-50% by aligning your proposal with what the funder actually funds, not just what they say they fund.

Step 2 Check-In

1. Proposal template library covers federal, foundation, and corporate grant types
2. Drafting workflow produces complete first drafts from grant announcements and client profiles
3. Needs assessments include specific data citations
4. Program descriptions align with evaluation criteria
5. Budget narratives justify every line item
6. Funder alignment analysis provides strategic positioning recommendations
7. Dual-model approach (Claude + GPT-4o) leverages each AI’s strengths

Step 3: Build the Compliance Checker

A single compliance error can disqualify an otherwise winning proposal. This module builds an automated compliance checking system that verifies every requirement before submission.

Sub-step 3a: Build the Compliance Checklist Generator

Create a Make.com scenario: “Compliance Check — [Client Name]”

1. Trigger: Proposal draft status changes to “Ready for Compliance Review”
2. Module 1 — OpenAI: Generate a compliance checklist from the grant announcement

Given this grant announcement, create a comprehensive submission compliance checklist. Include:
- Every required form and attachment
- Every formatting requirement (font, margins, spacing, page limits)
- Every content requirement (required sections, required information)
- Every deadline (registration, letter of intent, full proposal)
- Every certification and assurance required
- Eligibility verification items

Format as a checklist with checkboxes.

3. Module 2 — Notion: Create a compliance checklist page linked to the proposal

Sub-step 3b: Build the Automated Formatting Checker

Create a Google Apps Script that checks the proposal document for formatting compliance:

function checkFormatting(docId, requirements) {
  const doc = DocumentApp.openById(docId);
  const body = doc.getBody();
  
  const results = [];
  
  // Check font
  if (requirements.font) {
    const fontFamily = body.editAsText().getFontFamily(0);
    results.push({
      check: "Font",
      required: requirements.font,
      actual: fontFamily,
      pass: fontFamily === requirements.font
    });
  }
  
  // Check page count
  if (requirements.maxPages) {
    const pageCount = doc.getNumPages();
    results.push({
      check: "Page limit",
      required: "Max " + requirements.maxPages,
      actual: pageCount + " pages",
      pass: pageCount <= requirements.maxPages
    });
  }
  
  // Check margins
  // ... additional checks
  
  return results;
}

Run this script on every proposal before submission. It takes 5 seconds and catches errors that humans miss.

Sub-step 3c: Build the Pre-Submission Verification

Create a final verification checklist that you complete manually before every submission:

• All required sections present and in correct order
• Page limits respected (not one page over)
• Font and formatting match requirements
• Budget adds up correctly
• Budget narrative matches budget numbers
• All required attachments included
• Letters of support collected and signed
• Authorized signature obtained
• SAM.gov registration active (for federal grants)
• Submission portal access confirmed
• Submission deadline confirmed with time zone
• Proof of submission saved (screenshot or confirmation email)

Step 3 Check-In

1. Compliance checklist generated automatically from grant announcements
2. Formatting checker verifies font, page limits, and margins
3. Pre-submission verification checklist ensures nothing is missed
4. Proposals are submitted 24-48 hours before deadlines

Step 4: Build the Submission Tracking and Analytics System

Winning grants is a numbers game. The more proposals you submit, the more you win. This module builds the tracking system that manages your proposal pipeline and the analytics that prove your value to clients.

Sub-step 4a: Create the Submission Tracker

In Notion, create a database called “Submission Tracker” with these columns:

Sub-step 4b: Build the Analytics Dashboard

Create a Notion dashboard for each client with:

• Pipeline Funnel: Opportunities Identified → Applications Submitted → Awards Won
• Win Rate: Applications won / applications submitted
• Total Funding Secured: Cumulative grant awards
• ROI: Total funding secured / total fees paid
• Pipeline Value: Sum of all active proposals × win probability

Build a Make.com scenario that updates the dashboard monthly:

1. Trigger: Schedule — 3rd of each month
2. Module 1: Notion → Pull all submission data for the client
3. Module 2: Code by Zapier → Calculate all KPIs
4. Module 3: OpenAI → Generate narrative analysis
5. Module 4: Notion → Update the client dashboard
6. Module 5: Gmail → Send monthly report

Sub-step 4c: Build the Post-Decision Feedback Loop

When a proposal is declined, request reviewer feedback and analyze it:

1. Trigger: Submission status changes to “Declined”
2. Module 1 — Gmail: Send feedback request to the funder (if available)
3. Module 2 — OpenAI: When feedback is received, analyze it:

Given reviewer feedback on a declined grant proposal, analyze:
1. What were the specific weaknesses identified?
2. Which evaluation criteria scored lowest?
3. What did the winning proposals do differently?
4. What specific improvements should be made for future proposals to this funder?
5. Are there systemic issues in our approach that need to be addressed?

Provide a prioritized list of improvements for future proposals.

4. Module 3 — Notion: Add the feedback analysis to a “Lessons Learned” database

This feedback loop is your most valuable asset — it ensures every declined proposal makes your future proposals stronger.

Step 4 Check-In

1. Submission Tracker manages all proposals across all clients
2. Analytics Dashboard shows pipeline funnel, win rate, and ROI
3. Monthly reports are generated and delivered automatically
4. Post-decision feedback is analyzed and incorporated into future proposals
5. Lessons Learned database accumulates institutional knowledge

Step 5: Build the Post-Award Compliance and Reporting System

Winning the grant is only half the job. Managing the grant after award — submitting progress reports, tracking expenditures, documenting outcomes — is the other half. This module builds the post-award management system that extends your client relationship beyond the proposal phase.

Sub-step 5a: Create the Grant Management Tracker

In Notion, create a database: “Grant Management — [Client Name]”

Sub-step 5b: Build the Report Generation Workflow

Create a Make.com scenario: “Grant Report Generator”

1. Trigger: Schedule — 2 weeks before each report deadline
2. Module 1 — OpenAI: Draft the progress report

Given the grant's program description, the client's activities during the reporting period, and the funder's reporting requirements, draft a progress report that:
1. Summarizes activities during the reporting period
2. Reports on progress toward each objective (on track, ahead, behind)
3. Documents expenditures by category
4. Describes any challenges encountered and how they were addressed
5. Outlines planned activities for the next reporting period
6. Includes required metrics and outcomes data

3. Module 2 — Notion: Create the report draft
4. Module 3 — Gmail: Send the draft to the client for review and approval

Sub-step 5c: Build the Budget Tracking System

Create a Google Sheet: “Grant Budget Tracker — [Client Name]”

Track every expenditure against the approved budget:

Flag any category that exceeds 80% of its approved amount before the midpoint of the grant period.

Step 5 Check-In

1. Grant Management Tracker monitors all active grants
2. Report Generation workflow produces drafts 2 weeks before deadlines
3. Budget tracking flags overspending risks
4. Post-award management is offered as a monthly retainer upsell
5. Client dashboard includes both pre-award and post-award metrics

Pricing and Cost Breakdown

Service Tiers

Success Fee (all tiers): 3-5% of awarded grant amount

Cost Breakdown

Total monthly cost at 1 client (Growth tier): ~$162/mo Total monthly revenue at 1 client (Growth tier): $1,500/mo + per-proposal fees + success fees Total monthly cost at 5 clients: ~$350/mo Total monthly revenue at 5 clients: $7,500/mo + fees + success fees

Production Checklist

Before activating the grant writing pipeline for any client, verify every item:

• Client Profile completed with all required fields
• Grant Scanner finds relevant opportunities weekly
• AI scoring produces accurate match scores
• Proposal template library covers the client’s target funder types
• Drafting workflow produces complete first drafts
• Needs assessments include specific data citations
• Funder alignment analysis provides strategic positioning
• Compliance checklist is generated from each grant announcement
• Formatting checker verifies document compliance
• Pre-submission verification is completed before every submission
• Submission Tracker manages all proposals
• Analytics Dashboard shows pipeline funnel and win rate
• Post-decision feedback loop is active
• Post-award management system is ready (if client opts in)
• Success fee structure is documented in the client agreement