Automated Nonprofit Funding Opportunity Research and Outreach

## Problem Statement

The staff member at civic-dialog.org — a nonprofit focused on automation and AI education for other nonprofits — manually researches funding opportunities, identifies funder contacts, and writes individualized pitch emails from scratch. Each email takes approximately 2 hours, limiting outreach to 2–3 emails per week. The goal is to automate the discovery, research, and drafting steps so that volume can increase significantly without proportional increases in staff time.

---

## Current Process Steps

1. **Research (Manual Googling, ~60 min):** Staff manually searches the web for grants and funding opportunities relevant to nonprofits focused on AI/automation education and civic dialogue.
2. **Contact Identification (~20 min):** Staff visits the funder's website to find an appropriate contact person (name and email address) to reach out to.
3. **Opportunity Assessment (~20 min):** Staff reviews the opportunity details to determine whether civic-dialog.org is a good fit and identify the key alignment points to highlight.
4. **Email Drafting (~90 min):** Staff writes a tailored pitch email from scratch explaining what civic-dialog.org does and why it is a strong match for the specific funding opportunity.
5. **Email Sending (~10 min):** Staff sends the email directly to the funder contact.

---

## Integrations & Tools

| Purpose | Suggested Tool(s) |
|---|---|
| Grant/funding discovery | Candid (Foundation Directory Online) API, Instrumentl, Grants.gov, web scraping |
| Contact discovery | Hunter.io, Clearbit, or manual scraping of funder websites |
| AI email drafting | OpenAI GPT-4 or equivalent LLM via API |
| Email sending | Gmail API or SendGrid |
| Workflow orchestration | n8n, Make (Integromat), or custom Python pipeline |
| Data storage & tracking | Airtable or Google Sheets (to log opportunities, contacts, draft status, sent status) |

---

## Data Sources

- **civic-dialog.org organizational profile:** A structured document (to be created once) describing the organization's mission, programs, impact metrics, target population, and funding needs. This will serve as the base context for all AI-generated emails.
- **Funder opportunity listings:** Pulled from grant databases, funder websites, and search APIs. Should include: funder name, opportunity name, funding amount, eligibility criteria, focus areas, deadline, and URL.
- **Funder contact information:** Name, title, and email of the appropriate program officer or contact, sourced from the funder's website or contact discovery tools.
- **Sent email log:** A running record in Airtable or Google Sheets tracking which funders have been contacted, when, with which draft, and any responses received.

---

## Acceptance Criteria

1. **Discovery:** The system identifies at least 10 new, relevant funding opportunities per week from sources beyond manual Google searches, filtered to match civic-dialog.org's focus areas (AI education, automation, nonprofit capacity building, civic dialogue).
2. **Contact Identification:** For each discovered opportunity, the system automatically attempts to find the name and email of a relevant contact person at the funder organization.
3. **Fit Scoring:** Each opportunity is assessed and scored or flagged for relevance so the staff member can quickly prioritize which ones to pursue.
4. **Draft Generation:** For each approved opportunity, the system generates a complete, personalized pitch email that (a) accurately describes civic-dialog.org, (b) references specific details of the funding opportunity, and (c) articulates a clear alignment between the two. Drafts must be human-reviewable before sending.
5. **Human Review Gate:** No email is sent automatically without explicit staff approval of the draft. The workflow must include a review/approval step (e.g., a queue in Airtable or an approval email).
6. **Email Sending:** Upon approval, the email is sent from the staff member's email address (or a designated organizational address) to the identified contact.
7. **Logging:** Every opportunity researched, every draft generated, and every email sent is logged with timestamps and status in a central tracker.
8. **Throughput:** The automation should make it feasible for the staff member to send 10–15 outreach emails per week rather than 2–3, with no more than 30 minutes of human review time per email.

---

## Out of Scope

- Full grant application writing (lengthy proposals, budgets, attachments) — this automation covers letters of inquiry and initial outreach emails only.
- Automated follow-up sequences or CRM-style relationship management (may be a future phase).
- Automatic sending without human review — a human approval step is required before any email is dispatched.
- Managing inbound responses from funders.
- Applying to grants on online portals or forms.
- Financial tracking or grant reporting once funding is received.