Frequency capping sets a ceiling on exposures per user per window: three impressions a day, ten a week. Without one, a small well-defined audience will absorb an outsized share of budget through sheer repetition, and the advertiser pays for every one of those impressions.

How it works

The ad server tracks how many times each user has been shown a campaign and stops serving once the cap is reached until the window resets. Caps can be set per creative, per campaign, or across an account, and the level chosen matters: a per-creative cap on a campaign with eight creatives is barely a cap at all.

Why it matters

Ad response follows a curve that rises, plateaus, and then turns negative. Early exposures build recognition, middle ones convert, and past a certain point additional impressions stop working and start irritating. Spend continues either way. Capping keeps budget on the productive part of that curve and spreads it across more of the audience instead of deeper into the same people.

Where it matters most

• Retargeting. Retargeting is the usual offender because the audience is small and the budget is not.

• Small or narrow targeting. The tighter the segment, the faster the same people absorb the impressions.

• Long campaigns. Fatigue compounds over weeks in a way it does not over days.

• High-frequency placements. Formats a user encounters many times per session reach the cap far sooner than the media plan assumes.

Frequency capping and opt-in formats

The pressure is structurally lower in rewarded environments. A user opens an offerwall when they want to, browses a menu, and chooses what to engage with, so the format is not competing for attention that was pointed somewhere else. That is the difference opt-in makes: the exposure is requested, so it does not need rationing in the same way.

How to choose a cap

There is no universal number, but there is a reliable method: plot conversion rate against exposure count for a campaign that has been running without a cap. The curve almost always rises, flattens, and then declines, and the cap belongs near the point where it flattens. Considered purchases plateau later than impulse ones, and brand campaigns later than direct-response. Recalculate after significant creative changes, since a new execution resets fatigue and shifts the whole curve.

Common mistakes to avoid

• Setting one cap for every campaign. A considered purchase tolerates more exposures than an impulse one.

• Capping per creative and calling it capped. Users experience the brand, not your creative IDs.

• Ignoring cross-channel frequency. Platforms cap within themselves and cannot see each other.

• Capping so hard the message never lands. Too low a cap wastes reach by never completing the persuasion.