Average gift and median gift answer different fundraising questions. Average gift shows how much revenue each gift produces across the whole file. Median gift shows the middle transaction after all eligible gifts are sorted from smallest to largest. A few unusually large gifts can move the average sharply while leaving the median almost unchanged.
That is why a fundraising report should rarely show average gift by itself. The stronger standard is to publish the average and median together, then add the counts and concentration measures that explain the gap between them.
Average gift vs. median gift: the short answer
Average gift, also called the arithmetic mean, equals eligible gift revenue divided by the number of eligible gifts:
Average gift = total eligible gift revenue ÷ eligible gift count
Median gift is the middle gift after the eligible transactions are ordered by amount. If the list contains an even number of gifts, calculate the median as the average of the two middle values.
The UK Office for National Statistics describes the same general distinction: the mean uses the sum divided by the number of observations, while the median identifies the middle point and is less affected by a small number of extreme values. In fundraising terms, that means:
- Use the average when the question is about revenue yield per transaction.
- Use the median when the question is about the midpoint of transaction amounts.
- Use the distribution when the decision depends on gift bands, concentration, upgrades, or the effect of unusually large gifts.
One important wording rule: a transaction-level median describes the typical gift, not necessarily the typical donor. If one supporter gives monthly and another gives once, the transaction table gives those donors different numbers of observations.
A worked fundraising example
Suppose an appeal receives these nine gifts:
$25, $25, $50, $50, $50, $75, $100, $100, and $1,000
The gifts total $1,475. The average is $163.89 because $1,475 divided by nine equals $163.89. The median is $50 because the fifth value in the ordered list is $50.
| View | Revenue | Gift count | Average gift | Median gift |
|---|---|---|---|---|
| All eligible gifts | $1,475 | 9 | $163.89 | $50 |
| Illustrative view without the $1,000 gift | $475 | 8 | $59.38 | $50 |
The second row is not a recommendation to delete a valid gift. It is a diagnostic view showing sensitivity to one observation. The largest gift raises the average by $104.51, but the median stays at $50. Neither result is wrong. They describe different features of the same campaign.
If a leadership slide reported only the $163.89 average, someone might infer that a gift array centered near $150 reflects common donor behavior. The median warns that most transactions sit well below that amount. If the report showed only the $50 median, it would hide the real revenue contribution of the largest gift. The useful answer is the pair, not a contest between them.
Start with the reporting grain before calculating either metric
A metric can be mathematically correct and still answer the wrong question. Decide what one row represents before calculating average or median. This is the report’s reporting grain.
| One row represents | Metric you are describing | Useful question |
|---|---|---|
| One transaction | Gift amount | What did an individual donation look like? |
| One donor in a period | Donor-period value | How much did a supporter contribute during the period? |
| One campaign | Campaign-level result | What did the typical campaign produce? |
Do not label all three as “average gift.” A monthly donor who makes twelve $25 transactions has a $25 average transaction but $300 in annual donor value. Both numbers can matter, but they support different decisions.
The six-number gift-size profile
For a recurring campaign, channel, or period report, use this compact profile instead of one standalone average:
- Eligible revenue: the sum of gifts included under the report’s stated rules.
- Gift count: the number of included transactions.
- Donor count: the number of distinct matched donors, with unmatched or anonymous gifts disclosed.
- Average gift: eligible revenue divided by gift count.
- Median gift: the middle eligible transaction amount.
- Top-decile revenue share: the percentage of eligible revenue supplied by the largest 10% of gifts.
The first five figures establish scale and center. The sixth provides a simple concentration signal. When the average rises, the median stays flat, and the top-decile share increases, the likely interpretation is not broad-based gift growth. The change is concentrated toward the top of the distribution. That is a prompt to inspect gift bands and individual records, not a final diagnosis.
For a broader view of whether revenue depends too heavily on a narrow portion of the file, pair this profile with a donor mix stability report or a major gift volatility report.
What changes in the mean and median can tell you
| Pattern | Possible explanation to investigate | Next check |
|---|---|---|
| Average up; median flat | Larger gifts increased or one exceptional gift arrived | Top-decile share and gift-band revenue |
| Average flat; median up | Middle gifts improved while the upper tail weakened | Gift counts and revenue by band |
| Average up; median down | More small gifts arrived alongside stronger large gifts | New-donor volume, channel mix, and concentration |
| Average and median up | Gift sizes may have improved broadly | Counts, recurring mix, refunds, and comparable-period rules |
| Average up; gift count down | The remaining file may be larger-gift but narrower | Donor count, acquisition, retention, and the small-donor pipeline |
These are diagnostic patterns, not causal conclusions. A dashboard should make the next investigation easier without pretending the aggregate metrics prove why donor behavior changed.
Keep comparisons consistent across periods and channels
Average and median gift become misleading when the underlying inclusion rules change silently. Before comparing periods, campaigns, or channels, keep these choices stable or disclose the change:
- Status: Decide whether pending, failed, refunded, disputed, reversed, or adjusted gifts belong in the result.
- Date field: Use the same gift, settlement, deposit, or accounting date logic in both periods.
- Gift types: State how pledges, soft credits, in-kind gifts, grants, recurring installments, and matching gifts are handled.
- Currency: Convert consistently before aggregating multi-currency gifts.
- Channel and campaign mapping: Keep unknown and unmapped records visible rather than forcing them into a convenient bucket.
- Outlier policy: Do not remove a large gift simply because it changes the average. If leadership needs a recurring-performance view that excludes exceptional gifts, define that secondary view in advance and show both results.
When gift records can change after entry, the calculation should also follow the organization’s documented adjustment and close rules. The donation reconciliation report shows why transaction totals and deposited cash may differ without either being inherently incorrect.
How to place the metrics in a fundraising dashboard
A useful dashboard does not need a dense statistical panel. Start with one summary row containing eligible revenue, gifts, donors, average gift, median gift, and top-decile revenue share. Then provide three drill-downs:
- a gift-band distribution for volume and revenue;
- a trend view for average and median under the same filters; and
- a channel, campaign, and donor-status breakdown that preserves the underlying counts.
Always display the period, included statuses, transaction or donor grain, and last refresh time next to the metric. ReportWerks is designed to bring fundraising, marketing, and performance data into a connected reporting view; whichever reporting system you use, the value comes from keeping these definitions consistent and visible.
A five-check review before sharing the report
- Confirm that average gift uses eligible revenue and eligible gift count from the same filtered records.
- Sort the same gift population before calculating the median.
- Label transaction-level, donor-level, and campaign-level measures precisely.
- Inspect concentration and gift bands before explaining a change in the average.
- Run the final output through a fundraising report preflight so filters, calculations, comparisons, and decision ownership are clear.
The decision rule
Use average gift to describe revenue per transaction. Use median gift to describe the middle transaction. Use both, plus counts and concentration, whenever the report will influence ask strategy, campaign evaluation, donor-file health, or leadership expectations.
The goal is not to choose the more flattering number. It is to show enough of the distribution that the team can tell whether fundraising performance improved broadly, shifted toward larger gifts, attracted more entry-level donors, or became more dependent on a narrow group of transactions.
