Every impression served to the same person after a certain point costs money without producing a proportional return. The question is where that point is. Set the frequency cap too low, and the campaign will never build enough exposure to register with the audience. Set it too high, and the budget will keep paying for impressions that add nothing or, worse, irritate the viewer into ignoring the ad entirely. The right cap changes by format, audience, product, and objective, which makes it one of the more challenging variables to get right in campaign management.
In 1979, Michael Naples published Effective Frequency: The Relationship Between Frequency and Advertising Effectiveness through the Association of National Advertisers. The book compiled existing research on how many times a person needed to see an ad before it changed their behavior. Naples concluded that one exposure within a purchase cycle was usually not enough, two could have some effect, and three was the point where most of the impact had been delivered. Beyond three, diminishing returns set in.
That three-exposure threshold became standard practice in media planning for decades. Whether it still holds in a digital ad platform environment where impressions are cheaper and delivery is faster is a separate question, but the underlying principle has not changed: there is a point past which more impressions stop helping.
Where Excess Frequency Comes From
In an auction-based system, impressions go to the highest bidder. If a campaign’s targeting is narrow enough, the same users keep appearing in the auction pool. Without a cap, the system keeps buying those impressions because they match the targeting criteria, regardless of how many times the same person has already seen the ad. The result is a campaign that looks like it is delivering volume but is actually showing the same ad to the same people repeatedly.
A 2022 analysis commissioned by Google with Nielsen found that TV advertisers’ return on investment dropped by 41% when frequency exceeded six weekly impressions per viewer. That threshold accounted for 46% of all TV impressions served in the study. Nearly half of all impressions were past the point of diminishing returns.
Digital campaigns face the same dynamic, often faster, because impression costs are lower and delivery speed is higher. A campaign spending $3,000 a day on a narrow audience segment can reach oversaturation within 48 hours if no cap is set.
Frequency Caps and Reach Are Linked
Capping frequency does not just reduce waste. It redirects budget toward new users. When the system is told to stop showing ads to someone who has already seen the ad five times today, those unspent impressions go to someone who has not seen it at all. The campaign reaches more unique people for the same budget.
Google’s Display & Video 360 documentation on Added Reach Measurement confirms this directly. For some campaigns, the additional unique users reached because of a frequency cap accounted for more than 90% of total reach. That number depends on the campaign’s targeting breadth and the inventory available, but the principle is consistent: frequency caps convert wasted repetition into new reach.
TrafficJunky allows advertisers to set frequency caps at the campaign level, which means the cap applies across all placements within a campaign instead of per placement. This prevents a situation where a user sees the same ad three times on one site and three more times on another, hitting six total exposures even though each individual placement remained within its limit.
What Happens When Caps Are Too Tight
Setting a cap of one impression per user per day on a conversion-focused campaign is almost always too restrictive. Most purchase decisions, especially for higher-consideration products, require more than a single exposure. Naples’ research established this in the broadcast context, and digital data supports a similar pattern. The campaign needs enough impressions per person to move from awareness to action, and a cap of one rarely provides that.
The right approach is to test everything. Run the same campaign at different frequency settings, measure cost per conversion at each level, and find the point where the marginal cost of one additional impression per user exceeds the marginal revenue it produces. That point varies by product, by audience segment, and by creative. A cap that works for a direct-response campaign selling a subscription will not work for a brand awareness campaign promoting a new product. The only way to find the right number is to measure it.
