Recently Updated: 18 min read

Ad Frequency and Frequency Capping: How Often Per Person?

Ad Frequency and Frequency Capping: How Often Per Person?

Ad frequency is the number of times the same person sees your ad, and a frequency cap is an upper limit on that number. The official help pages do not name a correct number. They explain how a cap is set, what it counts and where it stops working, while the right figure depends on your goal, your creatives and the length of the flight. Buyers of ad space on publisher sites and publishers who sell it each need to ask a different set of questions. An online language course platform serves as the running example.

Frequency, reach and fatigue in plain terms

The first exposure introduces the brand, a second may help it stick, and after that each extra view tells the viewer less. When people see the same ad too many times it stops informing and starts to irritate. That effect is called ad fatigue. No official help page names the exposure count at which fatigue begins, so treat frequency as a variable to watch inside the campaign rather than a rule to look up.

Frequency matters for two reasons. The first is budget efficiency: impressions spent on people who have already seen the ad many times are impressions not spent on people who have not seen it at all. The second is experience: a reader who meets the same ad on every page may ignore it or install an ad blocker. How blockers affect publisher income is covered in ad blockers and publisher revenue.

Reach and frequency pull in opposite directions

Reach is the number of different people who saw the ad at least once, and frequency is the number of views per person. With a fixed budget, widening reach lowers frequency because impressions spread across more people, while raising frequency concentrates them on fewer. Which one you want depends on the goal. A new course launch wants reach. Moving hesitant visitors toward sign-up may justify a few more exposures per person. This is practical reasoning, not a rule from any platform.

The anatomy of a cap: count, window and level

A cap has a count and a time window, such as "no more than this many views per person per day". A low count means the ad seldom appears, while a high one makes the cap meaningless. A short window is strict, a long one is loose. Which windows can be chosen depends on the system.

Table: Cap features according to two help pages.

FeatureGoogle Ads (Display and Video)Ad Manager (line item)
PurposeLimit how many times ads appear to the same personLimit how many times a user is served a line item in a time period
Time windowPer day, week or month for Video. The page lists no interval for DisplayExamples on the page include hours, days, weeks, months and lifetime
Number of capsNo upper limit stated on the pageUp to 10 per line item
LevelDisplay: campaign, ad group or ad. Video: campaign onlyLine item. Creative and ad unit caps are separate
IdentifierDisplay uses third party cookies by default, first party cookies approximate impressions otherwiseA user identifier such as USERID or PPID is required for time based caps

Limits and options on these pages can change, so check the current versions before you rely on them.

The publisher's view: what Ad Manager lets you set

The Ad Manager help page on frequency caps (Opens in a new tab) describes the publisher side. You can add up to 10 frequency caps per line item. A cap is created in the "Adjust delivery" section by choosing "Set per user frequency cap", entering a maximum number of impressions and the interval over which the cap applies. When several caps exist, Ad Manager enforces all of them. The page gives examples with hours, days, weeks, months and lifetime caps.

The same page lists restrictions. Frequency caps cannot be set for line items in Preferred Deals. On roadblocked line items, a cap counts page views instead of individual ad impressions. Monthly and lifetime caps are not fully captured in forecasting, so forecasts can overpredict. A new Chrome session may show one extra impression past the cap because the first impression is used for cookie collection. On mobile, caps may be limited to a single domain or app when identifiers are restricted. Creative caps and ad unit caps are separate, and when both are set the more restrictive rule applies.

As a publisher, the cap is a tool that balances the advertiser's reach against your readers' patience. A very tight cap can affect delivery, because someone who has reached the limit no longer receives that line item and another ad takes the slot. Points worth discussing with an advertiser are who sets the cap, in which time unit, what happens for visitors without an identifier, and whether the report shows the effect. The documentation has no statement about other ad servers, so do not assume the same options elsewhere.

The advertiser's view: deciding whether to cap

As an advertiser you make the decision with three questions.

  1. Is the goal recognition, reminding hesitant visitors, or sign-ups?
  2. How many creatives do you have, and how long will the campaign run?
  3. Can the publisher apply a cap, and what happens to visitors without an identifier?

Table: Situation and suggested approach.

SituationSuggested approachNote
One creative, long flight, cap possibleSet a cap and monitor it in week oneTune the number with your own data, do not assume a ready made value
One creative, long flight, cap not possibleAdd creatives and shorten the flightRotation partly replaces a cap
Several creatives, short flightA cap may not be neededWatch for complaints and drops in click rate
Several publishersWrite a cap and an overall expectation per publisherPublishers do not see each other's impressions

The table is a practical suggestion, not text from a help page.

Stacking caps on one line item

The Ad Manager page allows up to 10 frequency caps per line item, set per hour, day, week or month, or as a lifetime total. When a line item has several caps, all of them are enforced. The page gives 1 impression per hour together with 3 per day as an example. If both a line item cap and a label cap on creatives or ad units exist, the more restrictive one applies.

SetupWhat it enforcesWhat to write in the plan
3 per day onlyThree impressions can still arrive back to backDecide whether back to back matters
1 per hour and 3 per dayBoth limits at onceList both limits
Line item cap plus label capThe more restrictive limit winsAsk which one binds

Where a cap stops applying

On mobile web, when third-party cookies are blocked, first-party cookies are used and caps apply only within a single domain. In mobile apps, device identifiers are used, a first-party ID may stand in if they are unavailable, and caps apply only within a single app. The documentation has no statement about cross-device capping. Practical suggestion: write the cap as "per browser or app", not "per person", so the client does not expect more than the system can deliver.

Forecasts and the Troubleshoot tab

Monthly and lifetime caps are not captured in forecasts, so a forecast can overpredict what will serve. The Troubleshoot tab of a line item shows reasons for non-delivery and whether caps are affecting pacing. Practical suggestion: when a capped booking under-delivers, open that tab before loosening the cap, and record what it showed. Menu names may differ if your interface is in another language.

Example scenario: a daily cap and a lifetime cap together

Example scenario: An online language course platform books a banner with 2 impressions per day and a lifetime cap for each visitor. The publisher confirms that the lifetime cap is enforced but absent from the forecast, so the platform plans for a lower delivery than the forecast suggests. It sets its internal expectations to the lower figure and checks the Troubleshoot tab after the first week.

A cap needs to recognize the same person. The Ad Manager page says time based frequency capping requires a user identifier, which could be a USERID or a PPID, and that without one the line item is not eligible to serve. Video pod and stream caps can serve without an identifier but then do not follow the cap. The Google Ads page says Display caps use third party cookies by default and use first party cookies to approximate impressions when third party cookies are not available.

For an advertiser the consequence is plain: setting a cap does not guarantee it works for every visitor. When consent is withheld or identifiers are restricted, the cap may cover a smaller audience or work only approximately. The effect of consent on measurement is described in cookie consent, privacy and measurement loss. Practical suggestion: ask the publisher how the cap treats visitors without an identifier and record the answer in the plan. Read every frequency figure as valid for the measurable audience, because the real frequency among visitors outside the cap may be higher than your count.

What Google Ads says about Display and Video caps

The Google Ads help page on frequency capping (Opens in a new tab) says capping limits how many times ads appear to the same person. It applies to Display and Video campaigns and is not supported in Demand Gen campaigns. In Display campaigns only viewable impressions count toward the cap. In Video campaigns a cap can be based on impressions, views or both, and impressions and views from other Video campaigns that use the same videos also count.

For Video the interval is per day, week or month, and the cap value must be a whole number, such as 4, not 4.5. For Display the page recommends letting Google optimize frequency, though you can set your own cap at the campaign, ad group or ad level. Display caps are set after the campaign is created, in "Frequency management" under campaign settings, and cannot be set during creation. Video caps can be set only at the campaign level. Once a person reaches a Video cap, Google stops showing that campaign to them, and the page says this applies only to in-stream and bumper ads in the auction.

These details describe Google Ads campaigns. A fixed fee slot on a publisher site is a different arrangement, and the settings may not be yours to change. The structure of the decision still carries over: a number, a window, a level and a way to recognize the person.

Rotating creatives before you cap

Think about creative rotation before setting any cap. If you rotate several creatives that share a message but differ in image, headline or call to action, the same person does not meet the identical ad each time. That reduces fatigue and shows which creative works. For testing methods, read the banner ad A/B testing sample plan, and for design choices, banner ad design tips.

A rotation schedule

Write the rotation as a small table: which creative goes live in which week, which one retires, and which metric movement brings in the next one. Putting this trigger on paper beforehand avoids the "let us wait a little longer" debate in the middle of a flight.

Sequencing messages across exposures

Lowering frequency is not the only lever. Showing a different message each time changes the experience even when the number of views stays the same. For the language platform, the first exposure can explain the teaching method, the next can show a sample lesson, and the last can state the trial terms. The documentation has no statement on whether a given publisher system can order creatives per person, so ask. If it cannot, split the creatives across weeks. The sequence then follows the calendar rather than the person, but it still eases fatigue.

StageMessageLanding pageutm_content
IntroduceHow the lessons are taughtMethod pagemethod
ConvinceA sample lessonFree lesson pagesample
DecideTrial terms and cancellationPricing pagetrial

Balancing exposure across several publishers

When one campaign runs on several publisher sites, each publisher counts only its own impressions. A reader can see the ad on two sites, and neither site's cap notices. The documentation has no statement about a shared cap across publishers. That means total exposure across publishers may not be controllable through per publisher caps alone.

Practical suggestion: keep the number of publishers small, ask about audience overlap and divide the budget with that overlap in mind. If you have no overlap data, add creatives and spread them differently across publishers so the chance of one person meeting the same ad twice drops.

Questions to put to each publisher

QuestionWhy it mattersIf there is no answer
Can you apply a frequency cap?Decides whether a cap existsLean on creative rotation
In what time unit and at what level?Sets how strict the cap isMonitor indirectly through the report
What happens to visitors without an identifier?Defines the audience the cap coversAssume the cap does not cover everyone
Can the report show the effect?Provides decision dataTrack indirect signals with your own tags

Reading the signals during the flight

You may not see impressions per person in a publisher report, so watch indirect signals: the ratio of impressions to reach, the trend of each creative's click rate and reader complaints. Alone each signal proves little, but together they support a decision. How clicks and views are counted is explained in measuring banner ad performance.

SignalWhat it may meanFirst response
Impressions rise, reach is flatThe same people are seeing moreAsk about the cap and identifier status
A creative's click rate falls over timeFatigue, or a change in audienceSwitch on the backup creative and track it separately
Complaints or blocker use riseThe reading experience may have sufferedAdd creatives and review flight length
High reach, low clicksMessage or placement, not frequencyReview creative and slot position

The word "may" is deliberate, since one signal can have several causes. A falling click rate can come from the season, a position change or the page content as much as from frequency. Viewability can distort the picture too, and the measurement standards are described in ad viewability and IAB standards.

Change one thing at a time

When you suspect a frequency issue, it is tempting to adjust several settings at once. If you tighten the cap, refresh the creative and move the slot together, you will not know which change helped. The language platform logs each change with a date in its plan and waits a few days before judging it. How long to wait depends on traffic volume, and no source gives a fixed duration, so with low traffic wait longer.

Example scenario: an online language course platform

Example scenario: An online language course platform runs a two week campaign on an education blog to bring back undecided visitors. It starts with a single creative. After two days the publisher's report suggests the same readers are seeing the ad repeatedly.

The platform first asks the publisher whether a cap is applied. The publisher uses Ad Manager and says it can add a time based cap to the line item, but visitors who decline cookies carry no identifier, so the cap does not work for them. The platform prepares three creatives, one on the teaching method, one on a sample lesson and one on trial terms, and rotates them weekly. Each carries its own utm_content value.

At the end, the platform looks at completed sign-ups as the primary metric and at reader complaints. It cannot say the cap alone raised sign-ups, because the creative change and the cap began on the same day. For the next campaign it decides to test the two changes separately.

When a cap is not the answer

Some frequency problems cannot be solved by a cap. If the creative is weak, a lower frequency just means fewer people see a weak ad. If the placement sits where readers never look, a perfect cap changes nothing about the result. If the landing page fails to deliver what the ad promised, extra reach only sends more people to a page that disappoints them. Before you tighten a cap, check three other things: whether the message matches the audience, whether the slot is seen and whether the page keeps the promise.

A cap also costs something. Each person who reaches the limit stops receiving the line item, so delivery volume may fall short of what you booked. If you pay by impressions, ask the publisher how unfilled impressions are handled, and if you pay a flat fee, ask whether the slot goes to another advertiser or to a house ad once a visitor hits the limit. These are questions about the contract, and the help pages do not answer them for every publisher, so put the answers in writing.

Checklist for the frequency decision

  • The goal is written: recognition, reminding or sign-ups.
  • The publisher was asked whether a cap can be applied.
  • The time unit and level of the cap are known.
  • The behavior for visitors without an identifier is known.
  • At least two creatives are ready, with a rotation calendar.
  • Each creative has its own utm_content value.
  • Someone owns watching complaints and click rate drops.
  • Cap and creative changes are tested separately, not together.

Misunderstandings to avoid

Setting a cap does not mean it works for everyone

Without an identifier the cap cannot apply. Writing "cap in place" in the plan is not enough, the plan must also say what happens when no identifier exists.

One system's rule does not hold for another

The 10 cap limit in Ad Manager, the Video intervals in Google Ads and the Display levels should not be mixed up. Each help page describes its own system.

Lower frequency is not always better

Since the help pages state no threshold, claiming that more than some number of views is bad would be wrong. Watch your own data against the goal and change the cap in small steps.

A cap does not cure creative fatigue alone

A cap reduces the number of times one person sees the ad, yet the remaining exposures may show the same creative. Plan rotation together with the cap and review both as the flight gets longer. If the flight is short enough that fatigue is unlikely, a cap may add little, and the time is better spent on creative quality and the landing page experience that follows the click.

Frequently Asked Questions

How many times should one person see my ad?

The help pages give no fixed number. The right figure depends on your goal, creative count and flight length. Start with a modest cap if the publisher supports one, watch your own data and adjust in small steps.

What does a Google Ads frequency cap count?

According to the help page, Display campaigns count only viewable impressions toward the cap. In Video campaigns a cap can be based on impressions, views or both.

How many frequency caps can an Ad Manager line item have?

The help page says up to 10 frequency caps can be added per line item, and Ad Manager enforces all of them.

Does a frequency cap work without a user identifier?

The Ad Manager page says time based frequency capping requires a user identifier such as a USERID or a PPID, and without one the line item is not eligible to serve. Video pod and stream caps can serve without an identifier but then do not follow the cap.

Can I limit total frequency across several publishers?

The documentation has no statement about a shared cap across publishers. Practical suggestion: keep the publisher count small, ask about audience overlap and increase the number of creatives.

What can I do if the publisher cannot apply a cap?

Rotate more creatives, shorten the campaign and write a refresh date in the plan. These measures do not fully replace a cap, but they reduce repeated exposure to the same ad.

Conclusion

A frequency cap puts an upper limit on how many times one person sees an ad, but the help pages do not name the right number. The Google Ads page explains how caps work for Display and Video, and the Ad Manager page explains up to 10 caps per line item and the need for a user identifier. As an advertiser, weigh your goal, creative count and the publisher's ability to apply the cap. As a publisher, tell the advertiser how the cap works and what happens to visitors without an identifier. Think about creative rotation before the cap, and record every change with a date.

When you are ready to build the campaign, browse publisher slots on the ad spaces page.

0 Comments

No comments yet for this article.

Post a comment

Please log in to leave a comment. Login