Nonprofit Automation — Phased

Nonprofit Automation — Phased
Nonprofit Automation From spreadsheets to smart workflows The Multiverse School · zoom into any topic FOUNDATIONS Get the AI working for you Setup & AI Basics • Create a ChatGPT or Claude account • Give the AI your org context: role, mission, programs, tone, partners • The rule: AI drafts, you approve — human-in-the-loop, always ✎ EXERCISE · Configure & Test — ~10 min Draft a donor thank-you email Summarize a set of board-meeting notes Analyze a program description for a grant AI Personas • Tell the AI to be a specialist: Grant Writer, Program Director, Fundraiser, Data Analyst • The persona sets the quality — 'act as an experienced grant writer…' beats generic • Build reusable templates: thank-you, grant narrative, board report, volunteer welcome • Iterate: draft → refine → refine → done ✎ EXERCISE · Prompt Library Sprint — 10 min Create 3 prompt templates you'll actually use at your org Save them in a doc for reuse Spreadsheet Analysis • Upload a CSV / spreadsheet to ChatGPT (Code Interpreter) or Claude • Analyze donor, program, or financial data; segment donors by behavior • Generate board-ready reports & charts ✎ EXERCISE · Your Data, Analyzed — 15 min Upload your data (or our sample set), then ask the AI to: 1) find the top 3 trends 2) flag concerns / anomalies 3) write a 1-paragraph summary for your board BUILD IT Ship a working automation Your First Automation • Make.com scenario: Gmail trigger → OpenAI (categorize) → Google Sheets (log) • AI extracts sender type, urgency, summary, and a draft reply • Free tier: 1,000 ops/mo · gpt-4o-mini ≈ $0.05/mo · total $0–5/mo • Test: email yourself, hit 'Run once,' watch it flow end-to-end ✎ EXERCISE · Build the 3-Module Flow — ~35 min, hands-on Create a Make.com account; add the Gmail trigger Add the OpenAI module (categorize + draft), then Google Sheets (log) Map the fields and test with a real email Email Triage System • Upgrade the flow: structured JSON (function calling) + a Router • Branch by type: Donor→Fundraiser, Volunteer→Coordinator, Grant→Grant Writer, General • Specialized agents per route, each with domain knowledge • Gmail DRAFTS only — you review before send; urgent items ping your team ✎ EXERCISE · Build Your Triage — 20 min Clone your first scenario, then add: 1) a Router with 3+ routes 2) at least one specialized agent 3) a Gmail Draft module on each route 4) Sheets logging on each route MAKE IT YOURS Build for your real org Build Your Own System • Pick a repetitive process at your org • Map it: trigger → decisions → actions → what gets tracked • Patterns: donor follow-up, volunteer onboarding, grant deadlines, event reg, board reports ✎ EXERCISE · Build Your Nonprofit System — 30 min Process map (5 min): trigger? decisions? actions? tracked? Build & test (20 min): trigger → AI extract → router → actions → logging Share (5 min): what you built + the time it saves What's Next • You leave with: configured AI, prompt library, analysis workflow, triage system, a custom automation • Free-tier reality: total $0–5/month • Going further: batch processing, CRM integrations, advanced routing, error handling • Resources: Make.com Academy · ChatGPT/Claude docs · the Multiverse community The Pattern Every automation is the same shape: Trigger → AI Extract → Route → Act → Log The AI thinks · the router decides · the actions do · you approve

Tour mode

Space / → next · ← previous
Home / End first / last
1…9 jump to frame

Explore mode

Click any highlighted region to zoom in.
Click further-in regions to drill deeper.
Backspace / Esc — zoom back out
O — return to overview

Anywhere

M — toggle mode
T — toggle frame list
F — fullscreen · R — restart
0 / 0 Tour