A donor list that counted the same people twice
A regional food bank preparing its year-end appeal finds that its donor database and its event sign-up sheets do not agree on who its supporters are.
Illustrative example. The organization and its data are invented to show a typical pattern. This is not a customer story.
What they hold
- A donor database built up over many years, with names, addresses, emails, and gift history.
- Spreadsheets from fundraising events, typed up by volunteers after each event.
- An export from an online giving page, with its own column names and date style.
The mess
One supporter, several records
A couple who give jointly appear as Mr. and Mrs. R. Alvarez in the database, as Rosa Alvarez on the giving page, and as Rosa and Miguel Alvares on an event sheet. Each record carries part of their history and none carries all of it.
Gift dates in three styles
The database stores dates month first, the giving page exports them year first, and volunteers wrote whatever came naturally. Sorted together, the gifts do not fall in order, and a few early-in-the-month dates could be read either way.
Placeholders in the address fields
Where an address was not known, staff over the years typed none, a dash, or the office's own address so the record would save. Those rows look complete and are not.
What it costs
- The appeal is mailed to the same household two or three times, which costs postage and, worse, tells a loyal supporter that the organization does not know who they are.
- Lifetime giving is split across records, so long-standing donors are not recognized as such and are left off the list for a personal thank-you.
- The board report overstates the number of active donors, and the following year's target is set against a figure that was never real.
- Mail sent to placeholder addresses comes back, and nobody can tell which returned envelopes mean a donor has moved.
What clean looks like
- Each household appears once, with its full giving history attached and a note of which records were combined.
- Every gift date is written the same way, and the handful of ambiguous ones were checked against bank deposits instead of guessed.
- Unknown addresses are empty, so they can be counted and followed up, and are left out of the mailing on purpose.
- The donor count in the board report can be traced to a list of people.
Problems in this example
- Near-duplicate recordsTwo rows describe the same real thing but differ enough that an exact comparison misses them.
- Name variantsThe same person or organization appears under several spellings, orders, and abbreviations.
- Mixed date formatsOne column holds dates written in several different styles.
- Ambiguous day and month orderA date such as 03/04/2025 could mean March 4 or April 3, and nothing in the cell says which.
- Disguised missing valuesA cell is not empty, but what it holds means there is no value: N/A, a dash, 999, or the word none.