/
/
Find bad emails and true duplicates in Bloomerang, not households
For nonprofit data teams who want to know which donor records to fix first, by asking their AI assistant to audit contact data.
On this page
Connector
Connect Bloomerang, pick your AI tool and paste the prompt.
The problem
Cleanup lists fill with households flagged as duplicates
Bounced emails, missing primary phones or addresses and duplicate constituents build up quietly in Bloomerang. Finding them means opening records one at a time, and shared addresses often turn out to be households, so cleanup lists fill with false alarms.
Without PopdockAI
- Open constituent records one at a time
- Check emails, phones and addresses by eye
- Flag shared addresses as duplicates
- Spend time on households that are fine
With PopdockAI
- Emails, phones and addresses are checked for each constituent
- Likely duplicates are grouped by surname and postal code or email
- Household members are left out of duplicate groups
- Issues are ranked critical to low with the top three fixes, nothing is changed
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Audit the contact data on constituents matching `{scope}` and list the records that need fixing, worst first. `{scope}` may be a surname, a city, or the word all to sweep everything.
**Step 1: Establish the record set**
If `{scope}` is all or empty, call `bloomerang-get-constituent` with `max-records` set to `{max-records}` (default 200) and no other filter. If it names a city, filter by city instead. Otherwise filter by `full-name.contains`.
Keep each record id, full name, status, type, primary email, primary phone and primary address fields.
**Step 2: Find bad email addresses**
Call `bloomerang-get-email-address` with bad set to true. Every account-id it returns that is also in your record set has an email that has bounced or been rejected.
**Step 3: Find missing and duplicated contact points**
For each constituent in the set, call `bloomerang-get-email-address`, `bloomerang-get-phone` and `bloomerang-get-address` filtered by account-id, and check for:
* no record at all of that kind
* records exist but none is marked primary
* more than one marked primary
* an address with no postal code, or no state
**Step 4: Find likely duplicates**
Within the record set, group constituents that share a surname and a primary address postal code, or that share a primary email. Each group of two or more is a candidate duplicate. Also call `bloomerang-get-household` and check whether the members of a group already sit in the same household, since that explains a shared address and is not a duplicate.
**Step 5: Report**
A table ranked by severity, one issue per row, giving constituent name, account number and the issue:
* critical: no usable contact point at all, or a bad primary email
* high: likely duplicate, or two records marked primary
* medium: no primary marked, missing postal code or state
* low: missing secondary contact details
Finish with counts by severity, and the three fixes that would clear the most records.
**Rules and Guidelines:**
* Read only. Repair nothing. Do not call `bloomerang-update-email-address`, `bloomerang-update-address`, `bloomerang-update-phone`, `bloomerang-update-constituent`, or any other tool whose name starts with bloomerang-add- or bloomerang-update-. Merging or editing constituent records is a decision for a person.
* Do not call a shared address a duplicate when the constituents belong to the same household.
* Do not treat an inactive constituent as a data problem. Report the status and leave them out of the severity counts.
**Validation:**
* If the constituent call returns exactly `{max-records}` records, say the sweep was truncated and state what it covered.
* If `{scope}` matches nothing, say so and stop rather than falling back to a full sweep unasked.Who it's for
One answer, three teams that use it
Database managers
Prioritise data cleanup
Issues ranked by severity and the three fixes that clear the most records give you a plan.
Audit all Bloomerang constituent data and give me the top three fixes.
Development officers
Reach donors at a working address
Bad emails and missing primary contacts are listed, so appeals reach the right place.
Which Bloomerang donors in Riverside have bad emails or no primary address?
Gift processors
Find true duplicates
Likely duplicates are grouped without household false alarms, ready to review and merge.
Find likely duplicate Bloomerang constituents with the surname Garcia.
How it works
What PopdockAI does when you ask
You name a surname, a city or everything, and your AI assistant audits Bloomerang constituent data. It tells you if the sweep was cut short.
01
Ask
You name a surname, a city, or ask for everything.
02
Check
It reviews emails, phones and addresses for each constituent.
03
Group
It finds likely duplicates and leaves out shared households.
04
Report
You get issues ranked by severity and the top three fixes.
What PopdockAI provides
The tools behind the answer
Ready made tools for Bloomerang and one secure connection to your AI tool. No code, no exports.
bloomerang-get-constituent
bloomerang-get-email-address
bloomerang-get-phone
bloomerang-get-address
bloomerang-get-household
This report is built from Get data tools only. PopdockAI applies the row, field and parameter rules your admin sets, and every query is written to the audit trail.
The outcome
Know which records to fix first, without false alarms
Related use cases
More ways to put this data to work
One prompt template with infinite possibilities.
Your AI assistant adapts its approach, tone, and capabilities to match each unique context while you maintain a single, manageable workflow.
Join forward-thinking companies that are transforming their operations with intelligent, secure AI automation.

.webp)
