Rules
Part of Nonprofit fundraising campaigns: the parts worth your attention
4 things nobody tells you about nonprofit fundraising campaigns benchmarks
Four benchmark traps quietly mislead nonprofit fundraising campaigns, including mismatched populations, stale periods, blended measures, and external figures.
What to take away
- A benchmark answers one question onlythe one its population, period and measure were built for.
- The Fundraising Effectiveness Project puts overall donor retention in the low 40s percent and new-donor retention near 20 percent. Neither describes your file.
- Rebuild any figure from your own donor data before it reaches a board deck.
- An external benchmark describes peers. Your internal threshold decides whether you act.
- Blended averages hide retention and churn. Split the file before you compare.
Four traps built into published benchmarks
- The population rarely matches yours. A sector benchmark pools organizations by mission, not by list size, gift range or donor age. The Fundraising Effectiveness Project, run by the Association of Fundraising Professionals Foundation with GivingTuesday, reports overall donor retention in the low 40s percent and new-donor retention near 20 percent.
If your mid-level donors give $500 and a study's median gift is $50, the comparison says nothing about your program.
- The period is usually older than the decision. Giving USA, researched at Indiana University's Lilly Family School of Philanthropy, reported $557.16 billion in US giving for 2023 in its 2024 edition. Treat that as historical. M+R Benchmarks covers the prior calendar year, so the edition year belongs in the citation.
A pre-2024 retention rate may predate a platform migration or a postal rate change. Which campaign parts actually move a decision is the subject of nonprofit fundraising campaigns: the parts worth your attention.
- The measure is often blended.Average gift mixes $20 one-time gifts with $5,000 major gifts. The mean looks healthy while the median and mode show what most donors actually do. The Blackbaud Institute's Charitable Giving Report tracks online giving separately, and online has held near a tenth of total giving in recent years. Ask which denominator a study used.
- An external benchmark is not a target.A peer's 42 percent retention rate is a description, not a goal. Your threshold comes from your acquisition cost, your average gift and your budget. Two organizations with the same retention rate can need opposite decisions.
Rebuild a benchmark from your own data
- Match the population. Filter your file to the same donor types the study counted.
- Match the period. Use the same length and season, not your best quarter.
- Match the measure. Recompute mean, median and mode as separate figures.
- Record the gap. If your rebuild lands far from the published number, the gap is the finding.
Rebuild the benchmark
- Sourcepublisher, year, method, sample size
- Populationeligibility, geography, segment, exclusions
- Periodstart, end, maturity, season
- Measurenet funds, donor quality, denominator
- Decisionwho acts, on what threshold
| Field | What to record | Test |
|---|---|---|
| Population | Donor type filter, list size | Does it match the study's cohort? |
| Period | Start and end dates | Same length and season? |
| Measure | Mean, median, mode | Which one drives the decision? |
| Source date | Report name, edition year | Is it still the current edition? |
Your donor database already holds the rebuild. Neon One, Bloomerang, DonorPerfect, Classy and Givebutter ship retention and average-gift dashboards. Candid publishes peer financial data to check against. The nonprofit fundraising campaigns checklist covers the fields teams most often forget to log.
Example: a food bank that set the wrong target
A mid-size food bank read that sector retention averaged 45 percent and set that as its target. Its own file said otherwise: most donors gave once at a community event and never returned, so retention sat near 22 percent.
The fix was not to chase 45 percent. It was to split the file into event donors, monthly givers and major donors, then set a separate threshold for each. Event donors got a reactivation sequence. Monthly givers got a churn watch. The blended number had hidden both problems.
Reactivation email in the United States must clear the FTC's CAN-SPAM compliance guide, which sets opt-out and identification rules. For that repair done in stages, see moving fundraising campaigns forward: small, staged wins that compound.
Internal thresholds, not peer averages
Your threshold is a decision rule, not a comparison. Write it plainly: if retention falls below X, change the stewardship sequence; if cost per dollar raised rises above Y, pause the channel. Fill both from your own program data.
For any benchmark you plan to test against your own campaign, the NIST experimental design guidance starts with the objective and the constraints. Descriptive reporting and a controlled effect estimate are different claims, and a benchmark supports only the first.
Before you repeat a vendor's claimed result in a board paper, the FTC guidance on soliciting and paying for online reviews covers fake feedback, selective requests and conditioned incentives.
A Canada note on the same numbers
Canadian readers work from a second set of sources. CanadaHelps publishes an annual report on Canadian giving, and the Canada Revenue Agency sets receipting and filing rules for registered charities.
Treat Canadian numbers as a separate set with their own population, period and currency. If your donors give in Canadian dollars, US figures from Giving USA will not describe them. Monthly giving programs in Canada: benchmarks smaller shops can use works through the size bands that matter for smaller files.
