← Back to glossary

Performance & Measurement

Incrementality Testing

Incrementality Testing

Incrementality Testing

4 min read

Incrementality testing measures the conversions your advertising truly caused — the ones that would not have happened without it — by comparing a group exposed to ads against a held-back control group. It cuts through attribution’s credit games to answer the only question that matters: did the ad actually add anything?

How incrementality testing works

How incrementality testing works

Split the audience. One group can see the ads; a control group cannot.

Run the campaign. Both groups are otherwise treated the same.

Compare conversions. The difference between them is the incremental lift.

Act on the truth. Fund what genuinely adds, cut what does not.

Ways to run it

Ways to run it

Geo tests. Hold back whole regions as the control.

Audience holdouts. Randomly exclude a share of users from ads.

Ghost ads. Log who would have been served, then compare.

PSA controls. Show a neutral ad to the control group.

Why it matters

Why it matters

Truth over credit. It measures cause, not just correlation.

Catches waste. It exposes spend that would have converted anyway.

Better decisions. It reallocates budget to genuine impact.

Challenges and considerations

Challenges and considerations

Scale needed. Reliable lift requires meaningful volume.

Complexity. Clean test design takes real discipline.

Opportunity cost. A holdout means some conversions go unserved.

Time. Lift can take weeks to read confidently.

Run this with your AI assistant — not another dashboard.

Run this with your AI assistant — not another dashboard.

OpenAds connects your assistant to every major ad platform, so you can plan, launch and optimise campaigns in plain language — approving the moves that matter.

Explore OpenAds