Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for crm managers

Donor Data Hygiene Agent

Find duplicate, outdated and incomplete donor records, propose fixes with evidence and keep every gift

Data cleanupSales
See how it worksGet access
Donor Data Hygiene Agent: what goes in, what the agent does and what you get

What it does

Duplicate and outdated donor records waste mailings and embarrass staff. The agent scans the database for likely duplicates, deceased flags, bad addresses and missing household links. It scores each match by name, address, email and gift history and proposes merges with the evidence. It checks every proposed merge to see that all gift history, notes and soft credits will be kept. Uncertain matches go to a review list rather than the merge list. After the coordinator approves, it applies the changes to a test copy and checks totals before the live update. Edge case: two donors named J. Smith share an address, but one is a parent and one a child, so the agent proposes a household link rather than a merge.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueApprovedNoNo 1 STARTS WHEN Monthly scan starts 2 USES A TOOL Export donor records and gift history 3 DOES Score likely duplicates and flag deceased and badaddresses 4 CHECKS THE RESULT Is each merge supported by at least two matchingfields? If not: Move uncertain matches to the review list. Backto step 2. 5 DOES Build the merge and fix proposals with evidence 6 YOU APPROVE Coordinator approves merges and fixes 7 USES A TOOL Apply approved changes to a test copy 8 CHECKS THE RESULT Do gift totals and counts match before and after? If not: Undo the merge and review the mismatch. Back tostep 7. 9 DOES Apply to the live database after approval 10 YOU APPROVE Coordinator approves the live update 11 RESULT Change log and updated database
Read the steps as a list
  1. Monthly scan starts
  2. Export donor records and gift history
  3. Score likely duplicates and flag deceased and bad addresses
  4. Is each merge supported by at least two matching fields?If not: Move uncertain matches to the review list. Back to step 2.
  5. Build the merge and fix proposals with evidence
  6. Coordinator approves merges and fixesThe agent waits here for your OK.
  7. Apply approved changes to a test copy
  8. Do gift totals and counts match before and after?If not: Undo the merge and review the mismatch. Back to step 7.
  9. Apply to the live database after approval
  10. Coordinator approves the live updateThe agent waits here for your OK.
  11. Change log and updated database

How it decides

Merges and fixes are proposed with evidence; no change is applied if gift history would be lost or the match is uncertain.

  • Require two matching fields for a merge
  • Never merge records with different gift histories without review
  • Mark deceased only with a source
  • Propose households for shared addresses with different first names

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Matching fields needed (default 2)
  • Scan frequency
  • Records excluded from merges
  • Backup rule

What keeps you in control

It always asks you first

  • Coordinator approves merges
  • Coordinator approves the live update

Hard limits

  • Never apply a change to the live database without approval
  • Never delete a gift record

It stops when

  • Done: approved changes applied with matching totals
  • Stop: totals do not match after a test

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensThe agent found 340 likely duplicates. It moved 41 with only a name match to the review list. In a test of 299 merges, total giving was $2,408,120 before and $2,406,320 after. The check found $1,800 in one merged pair lost. It undid that merge, fixed the gift link and the totals matched.

More agents for crm managers

Data cleanup · Quality control

Marketing Lead Data Repair Agent

Clean, deduplicated lead imports with a reversible repair log.

7 steps1 check4 jobs
Quality control · Sales

Dormant Deal Evidence Refresh Agent

A forecast that reflects real buyer evidence.

7 steps1 check8 jobs
Decisions · Sales

Sales Reference Matching Agent

Relevant, permitted references for every request.

8 steps2 checks2 jobs
Data cleanup · Sales

Sales Contract Data Handoff Agent

CRM records that match the signed order exactly.

7 steps1 check3 jobs