Careers
Part of Donor retention: the parts worth your attention
Habits that keep donor retention strong: what a business team should watch
Seven donor retention habits worth watching, from thank-you calls to monthly giving churn, with the reporting cadence that keeps each one honest.
What to take away
- Retention is a set of habits, not a campaign. The habits below are the ones a marketing lead can watch weekly without new software.
- Watch first-gift donors hardest. A donor who gives once and never again is the whole problem the habits exist to solve.
- Give every habit an owner and a cadence. A habit nobody owns is a wish.
- Read cohort renewal and reason-coded attrition together. A flat rate hides a growing lapse problem.
- Retire any routine that produces a report but never changes a decision.
The habits worth watching
Seven habits carry most of the retention result. None requires a new platform.
Seven donor retention habits
- Thank within 48 hours
- Call first-time donors
- Give the second gift a job
- Report back before you ask again
- Watch monthly giving churn
- Segment the lapsed list
- Keep ask and thank-you in same voice
Thank within 48 hours. A gift acknowledged inside two days renews at a higher rate than one thanked three weeks later. The habit is a queue with a daily owner, not a batch job at month end.
Call first-time donors. A phone call in the first month is the cheapest retention spend a small shop has. Assign names, not a list.
Give the second gift a job. A one-time donor who never sees a second ask has no path to a second gift. Decide what the next step is: monthly, a named program, a spring appeal.
Report back before you ask again. Donors who cannot say what their last gift did are being asked to trust a stranger twice.
Watch monthly giving churn. Recurring donors lapse quietly when a card expires. Failed-payment recovery is a retention habit, not a billing task.
Segment the lapsed list. A donor who gave for six years and stopped is not the same as a $25 one-time giver. One message to both is spam to one of them.
Keep the ask and the thank-you in the same voice. The person who signs the appeal should be recognizable in the receipt.
Cadence: what to review, and when
| Cadence | What to review | Decision it should force |
|---|---|---|
| Weekly | Thank-you queue, failed recurring payments | Who calls, who is re-contacted |
| Monthly | First-gift to second-gift conversion, reason-coded attrition | Which segment gets the next ask |
| Quarterly | Cohort renewal by entry year, cost per retained donor | Keep, change or retire a habit |
| Event driven | A lapsed-donor message that draws complaints | Pause, correct, then resume |
The weekly row is the one teams skip. It is also the only row that catches a donor whose card failed on Tuesday.
Review cadence and decisions
Weekly
- What to review
- Thank-you queue, failed payments
- Decision forced
- Who calls, who re-contacted
Monthly
- What to review
- First-to-second gift conversion
- Decision forced
- Which segment gets next ask
Quarterly
- What to review
- Cohort renewal, cost per donor
- Decision forced
- Keep, change or retire habit
Event driven
- What to review
- Lapsed message complaints
- Decision forced
- Pause, correct, then resume
Where AI touches these habits
AI now sits inside donor software: predicted lapse scores, send-time optimization, drafted thank-you notes. The NIST AI RMF Playbook lists voluntary actions under govern, map, measure and manage for teams putting a model into a workflow. It is a reference, not a certification, and it does not rank vendors.
The practical use is narrow. A lapse score can sort a call list. It cannot decide who deserves a call, and it should not write the thank-you a donor reads as personal.
Testing a habit without fooling yourself
Pick one habit, one population and one comparison basis. If you change the thank-you call script, compare donors who received the old script with donors who received the new one over the same period. For a fuller picture of what to watch after the test, see donor retention: the parts worth your attention.
The NIST experimental design selection guidance starts design choice with the objective and the practical constraints. That keeps descriptive reporting separate from a claim that the change caused the result.
Record the population, the dates, the version tested and the unresolved limits. The GAO data reliability guide treats data quality in relation to its intended use and calls for documented assessment. Your dataset is not certified by anyone; your team owns the facts and the decision.
If the test involves donor testimonials or reviews, the FTC guidance for marketers using reviews covers fake feedback, selective requests, conditioned incentives and hidden relationships. Quoting a donor who was paid, or only quoting happy donors, is the same problem in a fundraising context.
Worked example: a stalled spring appeal
A mid-size charity mails 12,000 lapsed donors in March. Response is 1.1%, and eleven complaints arrive about repeated contact.
The team stops the next drop, pulls the complaint names, and finds they were all contacted twice in the same week by two different lists. The fix is a suppression rule, not a new message. The next drop goes to 4,000 donors with no contact in 90 days.
Cohort renewal for the year is compared against the prior year, with cost per retained donor alongside it. The habit that changed was list hygiene, and it belonged to one named person.
Common questions
How often should a small team review retention?
Weekly for the thank-you queue and failed payments, monthly for conversion and attrition reasons. Quarterly is enough for cohort renewal and cost per retained donor.
What is the single strongest retention habit?
Thanking a first-time donor within 48 hours and calling them within the month. It is cheap, it is measurable, and it targets the donors most likely to lapse.
How do we know a habit is working?
Cohort renewal rises for the group the habit touched, and reason-coded attrition shows fewer "no contact" and "payment failed" exits. If neither moves, retire the habit.
Can software do this for us?
Software can queue the thank-you, flag a failed card and score a lapse risk. It cannot make the call, and a donor can tell the difference between a personal note and a template.
