Donor segmentation that actually changes what you do

A fundraising manager opens the donor database on a Monday morning. Down the left of the screen there are six labelled segments: major, mid, recurring, new, lapsed, prospects. The list is tidy. The call list for next week will look exactly the same as last week's, and the week before that. Nothing in those six labels has changed who is being called, what is being asked of them, or when. They are columns in a report wearing the word "segment" as a label.

This is what segmentation looks like in most Australian fundraising programs. The labels rarely change the actions a fundraiser takes; they describe a state of the donor base, in the language of a board pack rather than a call sheet. A segment that does not change what a fundraiser does on Tuesday is a slice for reporting, and labelling it as such would be more honest.

Segments that change next week's work

A segment is doing real work when a fundraiser does something different next week because of it. The segments that pass that test are the ones worth carrying.

Something like "donors at $1,000 to $5,000 last calendar year who gave within ten days of a peer-led appeal and have not been asked yet this year" passes. The fundraiser knows who to ring this week, what those donors responded to last time, and which bracket to ask in. The segment ends with someone dialling a number. "Mid-value donors" does not get a fundraiser anywhere; it tells a board what shape the donor base is in.

An action segment groups donors by what they are ready to be asked for next.

Why "size of last gift" is barely segmentation

Most Australian fundraising tools push organisations toward segmenting by last gift size, and sometimes by recency. That is two variables, and the smallest possible reading of a donor.

Donor behaviour carries far more signal than that. The channel a donor first came through, the appeal they responded to, whether they upgraded after the last ask, whether they have engaged with a peer fundraiser, what they wrote in the free-text field on the donation form, the cadence of their giving over the last three years: any one of these changes the calling list. The sector tends to ignore them because the underlying data is incomplete, or because the tool the team uses can only ask "size" and "recency" easily.

This shows up in the channels orgs default to. Australian fundraising still leans heavily on telefundraising and face to face. Both are blunt instruments, partly because nothing in the data tells the team how to be more precise. When the only thing the system reliably knows about a donor is the size of their last gift, the only available move is to call them and ask for more.

The action a segment carries

A useful segment carries the next action in its description. "Donors flagged for an upgrade ask next month." "Monthly givers due for a thank-you call in the next 30 days." "First-time donors who responded to a specific campaign and have not yet been asked for a second gift." Each is a list of people with a verb attached.

The biggest revenue most organisations leave on the table is the second gift that was never asked for. A segment that surfaces those donors in front of a fundraiser every week is what closes the gap. It does not need to be elaborate; it needs to land at the right time.

The recurring-giving segment behaves the same way. A list of monthly givers, narrowed to those whose gift has not been upgraded in 18 months and with a suggested new amount based on their first gift, is a call list. The point of a recurring donor over several one-off gifts is the compounding that comes from upgrades, and an action segment is where that compounding gets done.

Data that's right and stays right

Action-changing segmentation depends on data that is right and stays right. A segment called "lapsed donors in Victoria" is worthless if a third of those records are duplicates of active donors, if half the addresses are out of date, or if the lapsed flag was set six months ago and nothing has refreshed it since.

Most programs treat data hygiene as a quarterly project. A small team budgets a week to clean the database, the week stretches to a month, and by the end of it the records they cleaned at the start are already drifting again. The alternative is to keep the donor record right at the moment it arrives, and keep it right from there.

The reason this is hard in practice is that donor data is rarely in one place. A finance team member in one mid-sized Australian charity described spending hours each week trying to understand how a single donation was represented across Raisely, NationBuilder, and Stripe, only to find out later that a separate Zapier stack was feeding records in that nobody had documented. A segmentation layer built on top of that inherits every gap and every duplicate underneath it. Real segmentation starts with getting the donor record right as it arrives.

How many segments a team can actually act on

The real limit on action-changing segmentation is fundraiser time. A segment that drives action is a segment somebody has to act on. If the team can absorb three new weekly call lists, three new segments are useful; if it can absorb one, three is noise.

Many programs over-segment because they have a tool that segments well and nobody whose week changes as a result. The sensible move is to start with one new segment, attach it to one weekly fundraiser activity, and only add the second once the first has run for a quarter. Five new lists that nobody works through is segmentation back in its reporting clothes.

A starting place

A program that wants to test this approach can pick one segment with a clear next action attached. "Donors whose recurring gift has not been upgraded in 18 months" is a candidate. "First-time donors at $50 to $100 who responded to a peer-led appeal in the last 90 days" is another. Either produces a call list a fundraiser can work through this fortnight.

A quarter is long enough to see whether it landed. The fundraiser will know whether the list was worked through and what came back. A segment that earns its place shows up in the gifts that follow. One that does not quietly stops being looked at, and that is the cue to retire it.