Public Media Won't Stop Talking About Keeping Its Surge Donors. Why Is No One Talking About Finding the Next Ones?
Public radio just recorded its first decline in new donors since the 2025 rescission, and new-donor growth across the system has slipped from 66% to 50% year over year (Greater Public). None of this is a surprise, and the forecast called for exactly this: an inevitable plateau, then a likely decline, as the crisis that pushed a generation of first-time donors into station files settles back into something more ordinary.
The response has been a sector-wide turn toward retention, and it is the right instinct. Keep the surge donors. Convert the one-time givers before they lapse. Upgrade the monthly contributors who raised their gifts. Staff up major giving before the upgraded donors regress. We've spent the last several months making the case for the digital version of exactly this work: how to measure it honestly and why the pipeline runs through the membership function.
But notice what all of it assumes. Retention is a conversation about a funnel that is already full. The surge filled it, for free, on the strength of a once-in-a-generation news event. The harder question — the one almost nothing at the conference this summer will be built around — is what fills it next.
The whole room is planning to hold the line
The published guidance heading into this budget cycle is remarkably consistent, and remarkably good. Shift fundraising identity away from generic, crisis-driven drive messaging toward intentional, relational stewardship. Protect the upgraded donors, because 30% to 40% of those who stretched their giving in 2025 are projected to regress to prior levels within two years. Treat major gift staffing as a revenue strategy rather than a cost: Ideastream Public Media did exactly that, growing its $1,000-plus donor base 43% while pushing retention to 92%.
I won't argue with a word of it. Every station should be doing this.
The catch is arithmetic. Retention math only works if something keeps refilling the top of the funnel. A 92% retention rate on a file that has stopped growing is not stability — it is a slow, dignified decline, just one priced in five-year increments instead of all at once. The surge masked this for eighteen months because acquisition was happening on its own. Now it isn't, and the question of where the next cohort comes from is suddenly load-bearing again.
The question nobody's putting on a slide
When public media needs net-new donors at the top of the funnel, the historical answer has been to reach for a national model of who gives to public media: an aggregate donor profile pooled across the system, used to find prospects who resemble the people already giving at other stations. It is a reasonable-sounding idea, and it is also a direct-mail-era method wearing newer clothes. Find names that look like the names already on a file, somewhere.
That approach has three problems that the post-surge environment makes impossible to ignore.
It describes the average public media donor, not your market. A model built across the aggregate is, by construction, a portrait of the middle. It cannot tell you who in your coverage area is most likely to care about your newsroom, your local reporting, your music programming, the specific thing your station actually does. It hands you a generic resemblance when the entire advantage of public media is that the relationship is local.
It can only look backward. Any model trained on people who already gave is, definitionally, lagging. It finds more of who you already have. It is structurally incapable of surfacing the kind of donor the surge produced: someone who was never on anyone's giving file and showed up because the moment moved them.
It cannot see your most valuable prospect at all. The single best top-of-funnel candidate most stations have is the person who already loves the station and has simply never been asked in a way they could act on. I wrote last month about being that person: a loyal listener who tunes in from the car several mornings a week, has done so for years, and exists nowhere in the file. No national model will ever find me, because I'm not in any database of donors. I'm in the audience. That's a different list, and it's one you already own.
Public media's historical answer to top-of-funnel acquisition is to rent a generic model of who gives to public media nationally. The better answer is to put your own journalism in front of the people your own data says are most likely to engage with it.
You already own a better list
The mechanism is less exotic than it sounds, and most stations are already paying for the pieces.
Your content is the acquisition vehicle. Not a donation ask, but the actual work. The local investigation, the morning newscast, the session no commercial station would air. That is the thing prospective supporters already value and the thing the platforms are built to distribute to people likely to engage with it. You run it as top-of-funnel creative, and you seed the targeting not from a national profile but from your own first-party signal: the people who already stream you, read you, watch your video, open your email, and the donor file you already hold. The algorithm's job is to find more people who behave like them, in your market.
This does two distinct things at once, and last-click reporting will badly undercount both. It starts genuinely new digital relationships with people who would never have found their way to you otherwise. And through retargeting, it reaches the loyal-but-unfiled audience member, the car-radio listener, and finally gives them a door. One is acquisition in the truest sense. The other is converting an asset you already own and have never monetized. A national model does neither.
What Netflix and charity: water figured out first
This is not a public media insight, or even a fundraising one. It's the defining marketing shift of the last decade, visible anywhere a relationship is built at scale.
Netflix does not recommend your next show by buying a demographic profile of what people like you tend to watch. It watches what you watch. The recommendation works precisely because it is grounded in your own observed behavior rather than a purchased resemblance to a segment. A bought lookalike is a guess about a stranger; first-party behavior is knowledge about an actual person, and it consistently wins.
charity: water built one of the most recognizable donor bases in the nonprofit world without leaning on the traditional list-rental and list-swap economy. It produced its own content, told its own story, and owned the digital relationship directly, treating that first-party connection as the asset that compounds over time rather than something rented from someone else's file. The donors it acquired were never a borrowed audience. They were people who engaged with the work and were brought into a relationship around it.
The throughline across both, and across nearly every consumer business that has navigated the last ten years well, is a migration away from rented, third-party audience models toward first-party data, because what you observe about your own audience is more accurate, more durable, and cheaper to act on than anything you can buy about someone else's.
What acquisition on purpose looks like
None of this requires new technology. It requires deciding that acquisition is a discipline you run on purpose, with your own assets, rather than a thing you outsource to a model or wait for a crisis to do for you.
Treat your best content as your best acquisition asset. The piece of journalism or programming your audience already responds to is also your most effective prospecting creative. Put real distribution behind it, aimed at people who don't yet know you, and measure what it brings into the top of the funnel.
Build prospecting audiences from your own engagement, not a national profile. Your streamers, readers, video viewers, and existing donor file are the seed. Lookalikes built from your own first-party signal, in your own market, will out-target any aggregate model of the system because they're grounded in your actual audience rather than the average one.
Use retargeting to convert the audience you already have but never asked. The loyal listener who isn't in your file is the highest-percentage prospect you'll find and the one a national model is blind to. A standing digital presence is how you finally give that person a way in.
Measure it as pathway establishment, not last-click. Acquisition at the top of the funnel rarely converts on first contact, which means a last-click report will tell you it isn't working right up until the drive numbers say otherwise. Judge it by audience growth and pipeline, not by the final touch before a gift.
Finding is only the front of it
The donors you find this way sit at the front edge of an arc that runs far past acquisition. Once you've found them, you convert them into recurring relationships, you cultivate them through the membership function, and over the next decade you walk the right ones toward major and planned giving. Finding them is simply the stage the surge used to handle on its own, and the one the community’s current planning has the least to say about.
Find, convert, nurture, elevate: that sequence is the real shape of a donor relationship, and it's the lens worth carrying into a summer of conference sessions that will, by design, hand it to you one disconnected piece at a time. More on that to come.
