Constellations, patent pending

Public & Nonprofit · Process

Testing a Public Campaign

Before the money is committed, not after the year is over. The people who design an institutional message sit above and outside the people who receive it, and nobody in the approval chain is positioned to notice.

Quick answer

Institutional campaigns fail in a specific way. The people who design the message sit above and outside the people who receive it, the budget is committed before anyone checks, and the campaign runs for a year because that's what the funding cycle says. A perception test at the front costs a fraction of a print run and answers the only question that matters: does the audience read this the way the committee intended. For a university, a public agency or a grant-funded program, that check is cheaper than in commercial work and the consequence of skipping it is larger.

Why institutional messaging misses

Three conditions show up together in public-sector and nonprofit work, and each one makes the others worse.

The message is designed at a distance from the audience. A safety campaign written by people who don't do the job it's about. An enrollment campaign written by administrators rather than by anyone considering enrollment. A program launch written by the funder's communications team. The gap isn't incompetence, it's structural, and nobody in the approval chain is positioned to notice it.

The approval group is large and unrepresentative. More stakeholders than a commercial project, often with a value set or an agenda that doesn't match the population being addressed, and no one inside the group able to see that from where they're standing. The meetings are nominally collaborative. In practice the most senior or most forceful voice does the talking while the rest carry along.

The money commits early. Print runs, media buys, signage, event materials. In grant-funded work the budget is frequently locked at the proposal stage, which means the creative decisions get made against a number that can't move if the creative turns out to be wrong.

Figure 1 · Where the money commits

The funding cycle, with the budget locked at the start and the audience's read arriving at the end A timeline of five nodes: proposal, where the budget is locked; creative, where the direction is chosen in the room; production, the print run, media buy and signage; campaign, running for the funding year; and afterwards, when the audience's read arrives. A blue band marked money committed runs under the whole timeline. A green box at the front marks the check that works, before the direction exists, in 48 hours, at a fraction of a print run. A red box at the end marks the check that usually happens, after the year is over, with nothing left to move. PROPOSAL budget locked CREATIVE direction chosen in the room PRODUCTION print run, media buy, signage CAMPAIGN runs for the funding year AFTERWARDS the audience’s read arrives MONEY COMMITTED can’t move once the creative turns out to be wrong THE CHECK THAT WORKS before the direction exists 48 hours, a fraction of a print run THE CHECK THAT USUALLY HAPPENS after the year is over nothing left to move The budget locks at the first node. The audience’s read usually arrives at the last one.
The budget locks at the first node and the audience's verdict arrives at the last. Everything in between is spent against a message nobody outside the room has read.

Put those together and you get campaigns that satisfy the room and land badly outside it, discovered after the year is over.

The version that works

There's a well-known category of institutional message that gets built on the value the organization wishes it led with rather than the one its audience holds. The stated rationale is high-minded, the audience hears something they didn't ask for, and the campaign spends a year arguing with the people it needed.

The version that works usually says the same practical thing in the audience's own terms. It isn't a softer message. It's the same message pointed at what the audience already cares about instead of at what the committee wanted to be seen saying.

Which framing does that requires knowing something about the audience that nobody in the approval room has. It's answerable in advance, and it's cheap to answer.

What testing looks like on this kind of work

The method doesn't change for the sector. Audience perception mapping works the same way here as anywhere: an audience responds to images instead of describing them, and the pattern shows where the group agrees, where it splits, and why. What changes are the audiences and the timing.

Run it before the direction exists, not after. Collect a range of images and elements, put them to the population, and let the direction emerge from what they converged on. Bringing two finished directions to a test is too late, and it also gets a worse read, because a complete piece asks people to evaluate a whole rather than respond to the parts.

Segment by the groups most likely to disagree. For a university that's students, faculty, staff, alumni and prospective families. For a public program it's the population served, the frontline staff who deliver it, and the funders. For a workforce campaign it's the people doing the job against the people managing them. Where all segments align, that becomes the core. Where they diverge, that's a strategy decision made with evidence instead of assumed away, and what to do when the segments disagree is its own question.

Figure 2 · Segment by the groups most likely to disagree

A university campaign read by five segments across six elements, showing where they align and where they split A grid with five rows for students, faculty, staff, alumni and prospective families, and six columns for elements A to F. Each cell is a green dot for right for it, a red dot for wrong for it, or a dashed grey ring for no strong response. Elements A and B are green in every row and bracketed as the core, every segment aligned. Element D is mixed and bracketed as a strategy decision, the segments split. Element F is red in nearly every row and bracketed as out, rejected across the board. ELEMENT A ELEMENT B ELEMENT C ELEMENT D ELEMENT E ELEMENT F Students Faculty Staff Alumni Prospective families THE CORE every segment aligned A STRATEGY DECISION the segments split OUT rejected by nearly everyone right for it wrong for it The approval committee is one row that isn’t on this chart. The rows that are can’t be guessed from inside it.
Where every row agrees is the core of the campaign. Where the rows split is a decision, and now it's a decision made with the split in view rather than one assumed away in the approval room.

Use the internal audience. Institutions have a large, reachable, motivated population already inside the building, and it looks different in different institutions. At Express Newark it was the community partners and artists around a university arts hub, through a year of transition; in the college use case it's prospective students and the families deciding with them. Both are populations the institution can already reach, and both answered the question the approval room couldn't.

Test before and after where a program has a stated goal. For educational and behavioral programs, the same instrument run at both ends measures whether perception moved. That's a different use than creative testing and it's often the more valuable one on grant-funded work, because it produces something reportable.

Why the evidence matters more here

In commercial work, testing protects a budget. In institutional work it does that and one other thing.

Public and grant-funded organizations have to justify decisions to people who weren't in the room: boards, funders, oversight bodies, the population itself. A creative decision defended as “we felt this was the strongest direction” is weak in front of any of those groups, however good the direction is.

A decision defended as “we put a range of material to the people this serves, here is where they converged, and here is what we built on it” is a different kind of document. It survives a board meeting, a funder review and a change of leadership. It also makes late reversals harder, which matters on projects that run across fiscal years and staff turnover. Anyone who wants to move away from the audience's read now needs evidence to justify it.

Figure 3 · Two defenses of the same decision

The same creative decision defended two ways in front of a board, a funder and new leadership Two speech cards. On the left, in red, the usual defense: we felt this was the strongest direction. On the right, in green, the defense that holds: we put a range of material to the people this serves, here is where they converged, and here is what we built on it. Below them three rows, the board meeting, the funder review and a change of leadership, each with a red cross under the usual defense and a green check under the one that holds. THE USUAL DEFENSE “We felt this was the strongest direction.” THE DEFENSE THAT HOLDS “We put a range of material to the people this serves, here is where they converged, and here is what we built on it.” The board meeting The funder review A change of leadership weak in front of all three, however good the direction a different kind of document Same direction. Only one of these survives the people who weren’t in the room.
The direction can be identical in both columns. What differs is whether the decision survives the people who weren't in the room, and in institutional work there are always more of them.

The InEight rebrand is the enterprise version of the same effect: 22 stakeholders, 500 employees surveyed, one direction, zero revision rounds. The mechanism that produced that number is the one institutional work needs most, because institutional approval chains are longer and slower than commercial ones.

Fitting it into a funding cycle

The practical objection is timing, and it's usually smaller than it looks.

Results come back inside 48 hours. The work that takes real time is the part that happens anyway: understanding the program, the population and the goal, then assembling the elements worth putting in front of people. That's a normal discovery phase, not an added one.

Where it does need planning is the budget line. If the proposal locks creative production costs without a research line, add one, and size it against the print run rather than against the design fee. On most institutional campaigns the testing cost is smaller than the cost of reprinting a single asset, and dramatically smaller than a year of media pointed at the wrong framing.

Frequently asked questions

How can a nonprofit test a campaign before spending grant money?

Put a range of images and design elements to the population the program serves, segmented by the groups likely to disagree, before any direction is finalized. Results come back inside 48 hours, and the cost sits well below a single print run.

Who should be in the sample for an institutional campaign?

The population being addressed, the staff who deliver the work, and where relevant the funders and oversight groups. The approval committee is not a substitute for any of them.

Can perception testing measure whether a program changed anything?

Yes, by running the same instrument before and after. That produces a reportable movement in perception, which is often more useful to a funder than creative validation.

Why do public-sector campaigns miss their audience so often?

The message is designed at a distance from the people receiving it, the approval group is large and unrepresentative of them, and the budget commits before anyone checks. Each of those conditions makes the others harder to catch.