Ad pre-test

An ad pre-test evaluates a creative before it runs, on a battery that has barely changed in decades: likeability, emotion, message comprehension, intent, distinctiveness, and attribution to the brand. None of that changes when the respondents become synthetic. Only the clock does.

This is where synthetic quantitative pays off soonest, for a simple reason: pre-testing compares versions. And comparing — ranking, eliminating, ordering — is what a simulated panel is good at, even when it is poor at producing an accurate absolute level.

What reads, and what does not

Reads: the gaps between executions, the objections in the open-ends, where comprehension breaks, which version wins and by how much. Does not read: an absolute likeability score benchmarked against a market norm, or a projection into share. Synthetic steers a choice between creatives; it does not stand in for the measurement that releases a media budget.

What a buyer can check

That the audience is calibrated on the market the ad will run in. That the open-ends are not uniformly warm — no creative has only strengths. And that the reported gaps survive a second run.

For the wider practice, testing headlines, claims and body copy as well as finished ads, see copy testing.