Fundraising has a vocabulary problem that turns into a measurement problem, which then turns into a strategy problem. First we use a word too loosely, then we build reports around it, then we manage to the report as if the word meant what we hoped it meant.

“Acquisition” is one of those words.

In most fundraising reports, acquisition means the donor is new to the file. That is administratively true and strategically incomplete. A donor can be new to your CRM and still come from the same overworked, over-mailed, over-modeled pool every other organization is chasing.

New to file is not the same as new to market.

A lot of what gets counted as acquisition spend is not audience expansion, it’s buying new records from the same donor-response universe: same co-op, same list broker, same modeled charitable responders, same current-donor lookalikes, same channel, same offer, same assumption about who “our donor” is.

Then, after doing roughly the same thing for the 847th time, everyone stares at the campaign report and wonders why costs are up, response is down, the file is aging, and growth feels like trying to parallel park a refrigerator.

Replenishing the database with people who look like the last people can still produce donors and be economically rational but it’s not  audience expansion.

The KPI we’re missing

Acquisition metrics tend to the myopic – cost per donor,  response rate, average gift.  Those are useful, but are mostly telling us if harvested efficiently. They don’t tell us whether we’re reaching materially new people.

That is the missing KPI:

Audience Expansion Spend %

Or, more wordy but preceise:

What percentage of our spend is actually aimed at people we are not already chasing?

A working definition:

Audience Expansion Spend % is the percentage of fundraising and marketing spend intentionally aimed at reaching people outside the organization’s existing donor file, legacy list universe, co-op continuation pool, current-donor lookalike model, or repeatedly harvested audience source.

That includes genuinely new audience pools, adjacent identity groups, broader reach media, community distribution, content, influencers, partnerships, door, street, and other channels where the organization is getting outside the usual donor-response machine.

It does not include calling every Meta campaign “acquisition” because the people have not given yet. It does not include branded search. It does not include retargeting nor renting the same names while pretending the list broker has discovered some untouched cave of philanthropic unicorns.

The direct mail problem is not direct mail

The argument is not “direct mail bad.” Direct mail is a channel, not a moral failing.  The problem is same-pool dependency.

If you have mailed the same modeled universe for years, and the “new donor” names keep coming from the same co-op logic, same list sources, same continuation names, same demographic profile, and same charitable-response history, then the spend may be acquisition by CRM definition but not by growth definition.

If the whole market is pulling from the same pool, the best names get more expensive, more solicited, more irritated, and less responsive. The model says they are good names because their past behavior says they respond to charity offers. But that same past behavior is why everyone keeps burying them in asks.

Everyone acts rationally in the short term as the co-op gives you the highest-scoring names. Nobody wants to add names that might make the next report look worse.  And so the shared donor pool gets mined harder and harder. The spreadsheet says we are being disciplined and the market says we’re eating the seed corn.

The old list world knew this, awkwardly

Back when even unheard-of brands could mail millions of pieces a year and make money on acquisition, the people running the data still knew they needed new names.  Even when the machine worked, they understood the pool needed oxygen.

So they would sometimes add 10 to 15 percent of names from the bottom of the model into the mix, these are names that did not look quite as juicy in the short-term math.   The co-ops would sometimes do this without telling the client.  Maybe that sounds sneaky and maybe it was but the logic wasn’t stupid. They were trying to prevent the commons from collapsing.

Fast-forward to today and the need for new names is dialed to 11 on the panic meter but now acquisition is a loss leader and the microscope of cost per name dominates.  In that environment, no vendor wants to say, “We intentionally added lower-modeled names because the ecosystem needs replenishment and your future growth requires some controlled inefficiency.”

The irony is that some of those “less good” names may be less good because they are less exposed. They have not been trained by fifty-seven organizations to recognize and ignore the same urgency cue, the same premium, the same sad-eyed photo, the same faux-personal letter, the same “Dear Friend” wearing a fake mustache.

Lower model score does not automatically mean low potential. Sometimes it means less strip-mined.

have not been trained by fifty-seven organizations to recognize and ignore the same urgency cue, the same premium, the same sad-eyed photo, the same faux-personal letter, and the same “Dear Friend” wearing a fake mustache.

Lower model score does not automatically mean low potential. Sometimes it means less strip-mined.

And digital is not some magical escape hatch from this problem. In some ways, it’s worse. At least in the co-op world, a name and household usually correspond to a human being who lives somewhere real. The digital ad ecosystem is more like a casino designed by surveillance engineers and raccoons. It is full of bots, fraud, accidental clicks, muddy attribution, platform-defined audiences, black-box optimization, and algorithms whose basic sales pitch is: insert credit card here and trust us.

So when a digital campaign says it is “acquiring new donors,” the first question should be the same one we ask of mail: new compared to what? New to the file? New to the audience universe? New to the cause? Or merely new to this week’s reporting spreadsheet?

Every organization says it wants growth. But a lot of acquisition reporting punishes the behavior required to create growth.

Growth requires exploration. It requires some controlled inefficiency. It requires brand and reach work that does not convert tomorrow morning at 9:03. It requires audience learning, not just audience extraction.

But the dominant reporting culture rewards short-term harvesting. It asks, “What was the cost per donor?” before asking, “Did we expand the audience?” It asks, “Did this list beat the continuation?” before asking, “Are we too dependent on continuation names?” It asks, “Did the campaign pay back?” before asking, “Are we building a future market or liquidating the current one?”

Efficiency is good when the machine is pointed at the right target. But efficiency applied to a shrinking audience is just a more elegant decline curve.

The answer is not to abandon acquisition metrics. The answer is to stop pretending they are sufficient.

At minimum, organizations should split acquisition spend into different strategic categories.

Same-pool harvesting is spend aimed at the same repeatedly used donor-response universe: continuation lists, legacy co-op models, current-donor lookalikes, branded search, retargeting, and other sources that mostly capture people already close to the category or already inside the sector’s shared machinery.

Pool replenishment is spend intentionally allocated to lower-exposure, adjacent, under-modeled, or newly introduced prospect pools. These names may not look as efficient in the next report, which is precisely why they need to be protected as a separate test budget instead of shoved into the same short-term grading system.

Net-new reach is spend aimed at people outside the usual acquisition universe. This can include door, street, broader paid media, content distribution, partnerships, influencer activity, community presence, and other channels where the organization is trying to become known, salient, and relevant to people who were not already sitting at the bottom of the funnel waiting to be converted.

That is a better conversation than staring at cost per donor like it descended from Mount Sinai holding a calculator.

Kevin