How we measure donor behaviour
DonorCARE scores every donor on how they actually give, not on how many rows their receipts happen to produce — and that single choice changes what every number on a donor record means.
Why it works this way
Section titled “Why it works this way”A donor does not give in database rows. One visit to a ceremony can produce a whole basket of tablets, and one admin sitting down with a stack of paper receipts can key in a dozen donations in one afternoon. If DonorCARE counted every one of those rows as a separate “gift”, a generous once-a-quarter visitor who buys ten tablets in a single sitting would look, on paper, like ten separate acts of giving packed into one day — and the maths that is supposed to measure loyalty would instead be measuring basket size.
So DonorCARE counts giving days instead of donation rows wherever it measures behaviour: how often someone gives, how regular they are, and whether they are overdue. How much lifetime value they have added up is still a row-based total — that part is unaffected.
The terms
Section titled “The terms”Giving day — one calendar day (Malaysia time) on which a donor made at least one completed donation, carrying that whole day’s total. Ten tablets bought in one visit are one giving day worth the basket total, not ten gifts of a tenth each.
Cadence / effective interval — the gap DonorCARE expects between a donor’s giving days. It uses the donor’s own average gap once they have given on at least three separate days; below that it falls back to your organisation’s typical gap, and if your organisation doesn’t yet have enough dated history to know that either, it falls back to a fixed default. This is why a donor’s score doesn’t simply reset to zero the day after “average” — the bar it’s held to is their own rhythm, or your organisation’s, never a single fixed number.
Overdue ratio — days since a donor’s last gift, divided by their effective interval. This is the number that actually drives the recency part of the score, and it is the whole point of the scheme: a donor who typically gives once a year and is 200 days past their last gift is on schedule, not lapsing, because 200 days is well inside their own yearly rhythm. A ratio at or under 0.5 scores as fully on-time; by three times the expected interval, it scores as effectively lapsed.
Confidence — how much giving history stands behind a score, based on the number of giving days on record. It’s shown as a fade on a low-confidence score, not as a separate number — a donor seen on only one or two occasions gets a score, but it’s shown as tentative.
Churn risk — a Low / Medium / High / Critical read on the same overdue ratio, with a floor and some overrides: a donor with a healthy active recurring gift is never rated above Low, while a failing or recently-lapsed subscription, a declining giving trend, or a recent refund or void can each push the rating up.
What this means in practice
Section titled “What this means in practice”Do
- Read “days since last gift” next to the donor’s own typical gap (or use the Risk column, which already does this comparison for you) before deciding someone has gone quiet.
Don’t
- Assume a big gap in days always means a lapsing donor — a once-a-year temple donor and a monthly donor need completely different gaps before either one is actually overdue.
If your organisation imported years of paper records in one batch and stamped them all with the same entry date, DonorCARE has no way to tell those giving occasions apart — they collapse into a single giving day no matter how many separate visits they really were. Cadence and overdue numbers for those donors are only as good as the dates behind them; they become reliable again as new, individually-dated gifts come in.
Worked example
Section titled “Worked example”Ravi (an invented example) visits a temple three times a year and buys a small basket of tablets each visit:
| Visit | Day | Tablets bought | Rows | Giving day total |
|---|---|---|---|---|
| 1 | Day 0 | 8 | 8 | RM320 |
| 2 | Day 120 | 3 | 3 | RM150 |
| 3 | Day 240 | 5 | 5 | RM250 |
That’s 16 donation rows, but 3 giving days — the unit DonorCARE actually scores on. Ravi’s own gap between giving days is 120 days (the span from day 0 to day 240, over two gaps). Sixty days after his last visit, his overdue ratio is 60 ÷ 120 = 0.5 — right at the top of the scale, so his recency score reads as fully on-time.
Counted by row instead, those same 16 rows spread over 240 days would suggest a gift roughly every 16 days — and by that (wrong) yardstick, Ravi would already look badly overdue at 60 days out. Counting giving days is what keeps a basket of tablets from being mistaken for infrequent giving.
Related
Section titled “Related”Screenshots use sample data. Every screen on this page is captured against a demo organisation seeded with generated donors, amounts, receipt numbers and campaigns. Any resemblance to a real person or organisation is coincidental.