What is frequency capping? Frequency capping is a limit on how many times one person can see the same ad within a set time window, like three impressions per user per day or ten per week. It exists to curb two problems at once: wasted spend on people who have already seen the ad enough, and ad fatigue, where over-exposure turns interest into irritation. Set it well, and your budget reaches more people instead of hammering the same few.
If you have ever seen the exact same ad chase you across every video, every site, and every app for a week straight, you have met an advertiser who skipped this setting. Frequency capping is the fix. It is also one of the least glamorous and most quietly important levers in paid media, because getting it wrong costs you money on both ends: you pay for impressions that do nothing, and you train people to resent your brand.
What is frequency capping, and how does it work?
A frequency cap is two numbers and a clock: a count (how many impressions) over a time window (per day, per week, per campaign flight). "Three per day" means once a single user hits three impressions of that ad, the ad platform stops serving it to them until the window resets.
The cap is enforced at the user or device level, which is the part that matters. To say "this is the same person we showed the ad to yesterday," the platform needs an identifier: a cookie, a mobile advertising ID, a logged-in account, or a household IP. The cap is only as accurate as that identity signal. When the signal is clean, capping is precise. When it is fragmented, the cap leaks, and people see your ad far more often than your settings claim. This is the same identity problem that underpins most of programmatic, which is why the demand-side platform you buy through, and how it dedupes users, has a direct effect on whether your cap holds.
You can cap at different levels depending on the goal:
- Per ad (creative): stops a single creative from burning out.
- Per campaign: controls total exposure across all the creatives in one campaign.
- Per channel or cross-channel: limits how often a person sees you across display, video, social, and CTV combined. This is the hardest version and the one most advertisers never achieve.
A quick clarification, because the terms get muddled: frequency capping limits exposures, viewability measures whether those exposures were on-screen long enough to count, and CPM is what each thousand impressions cost you. A cap controls how often; viewability controls whether it landed; CPM controls what it cost. All three feed the same question: is this impression worth serving again?
Why frequency capping matters
Two reasons, and both are about money.
It stops waste. Ad response is not linear. The first few exposures do most of the work. After a point, each additional impression on the same person delivers less and less, while still costing you the same per impression. Without a cap, a slice of your budget gets spent re-serving the people who were already going to convert or already decided not to. A cap reallocates that spend toward reach, so more distinct people see you. If you want to see this in the harder numbers, it is closely tied to incrementality: the over-served impression is the textbook example of spend that takes credit for conversions it did not cause.
It prevents ad fatigue. Over-exposure does not just stop working, it actively backfires. The same creative shown too many times stops registering, then starts annoying. That irritation attaches to the brand, not the ad platform. Frequency capping is the guardrail that keeps a campaign on the productive side of that curve.
The honest version: there is no universal "correct" cap. The right number depends on the channel, the creative, the funnel stage, and how long your buying cycle is. A retargeting ad to a warm visitor tolerates more frequency than a cold prospecting ad. The job is to set a sane starting cap, then read the data and adjust, not to set it once and forget it.
What a "good" frequency cap looks like
There is no magic number, and any agency that hands you one without asking what you are running should make you nervous. That said, common starting points fall in sane ranges: roughly 3 impressions per user per day or somewhere in the high single digits to low teens per week for display and video, then tightened or loosened based on performance.
The right cap moves with context:
- Funnel stage: prospecting (cold) usually caps tighter; retargeting (warm) can run higher because intent is already there.
- Creative volume: more creative variations let you raise total frequency without fatigue, because no single ad gets overplayed.
- Sales cycle length: a considered, long-cycle B2B purchase tolerates more touches over time than an impulse buy.
- Channel: a 30-second CTV spot you cannot skip wears out faster than a small banner.
The method beats the number. Set a starting cap, watch frequency reports against conversion and engagement, and adjust. Rising frequency with falling response is the signal you have gone past useful. When you are pressure-testing a budget against reach and frequency assumptions, our ad budget calculator is a quick way to see how the math shakes out before you commit spend.
Frequency capping on CTV (this is where it gets hard)
On classic web and mobile, frequency capping is mostly a solved problem, because there is usually a stable identifier to count against. Connected TV is the channel where it quietly breaks.
The problem is identity. A household streams across multiple apps, multiple devices (smart TV, streaming stick, phone, tablet), and often without a single logged-in identity that ties them together. Each app or supply source may only know its own slice of exposures. So you can set a clean "3 per day" cap and still have a viewer hit your ad five or six times, because no single system has the full count across every app they watched. The more apps and supply paths your buy touches, the wider that gap tends to open, which is one reason a tightly managed CTV buy beats spraying impressions across every available source.
This is cross-device, cross-publisher frequency, and it is genuinely difficult on CTV without an identity layer stitching the exposures together. Approaches that help include household-level identity graphs, IP-based household matching, and signals like automatic content recognition, which can help reconcile exposure across screens. None of them are perfect, and anyone claiming flawless cross-app CTV capping is overselling. The realistic goal is managed frequency, not a guarantee, and knowing the difference is half the skill.
If you are weighing CTV at all, two pages are worth a read alongside this one: our guide to OTT and CTV advertising covers how the channel buys, and how much CTV advertising costs sets realistic expectations on spend before frequency ever enters the picture.
Frequency capping vs. sequential messaging
Frequency capping limits how often someone sees an ad. Sequential messaging controls what they see in what order, deliberately walking a person through a planned series of ads (awareness, then proof, then offer) across exposures.
They work together. Capping keeps any one message from being overplayed; sequencing makes the repeated exposures go somewhere instead of looping the same spot. Frequency capping is also the guardrail that keeps retargeting from tipping into harassment: the over-serving guard that stops "we noticed you" from becoming "we will not leave you alone."
How we use it
At MoonSauce, frequency capping is a default, not an afterthought. Across programmatic, CTV, and retargeting, we set caps deliberately, report frequency alongside the metrics that matter, and adjust as the data comes in. On CTV specifically, we are straight with you about what cross-app capping can and cannot promise, because the honest answer is more useful than a confident one that does not hold up. No vanity dashboards. No paying to annoy your future customers.
Want media that respects the budget and the audience?
Frequency capping is one small setting, and it is a decent tell. Agencies that ignore it tend to ignore the rest of the unglamorous work that protects your spend. We do not. If you want programmatic, CTV, and retargeting run by senior people who set the caps on purpose and tell you the truth about what CTV can and cannot promise, let's talk. No pressure, just a real conversation, just real talk. Email admin@moonsauceagency.com or book 30 minutes.
Keep reading: What is automatic content recognition (ACR)? · What is retargeting? · What is incrementality? · Programmatic advertising services · Back to the glossary
