Limited ad serving is the most consequential Google Ads policy most advertisers have never read, and this month it expanded to cover everything. It does not disapprove your ads. It reduces how often they appear, in scenarios Google does not enumerate, and it tells you almost nothing about why.
If a campaign has looked inexplicably soft since the summer, this is the first thing worth checking.
What actually changed this month
Google’s August 2026 policy update, posted on August 5, is three sentences long and carries more weight than its length suggests.
The first: "In August 2026, Google will update its Limited Ad Serving policy to cover all Google Ads." Not Search, not YouTube, not a product family. All of it.
The second is the timeline: "Implementation will begin gradually and will be completed by 2028."
The third is the mechanism: "Ads from unqualified advertisers will be limited in specific scenarios to decrease the potential for negative user experiences."
Read that last sentence carefully, because every important word in it is doing work. Limited, not rejected. Unqualified, a status rather than a violation. Specific scenarios, which are not listed.
Limited ad serving started in September 2024 and ends in 2028
The single most useful thing to understand about limited ad serving is that it is not a policy launch. It is the visible middle of a four-year rollout, and almost no coverage has put the dates end to end.
The policy itself records that "Initial enforcement for this policy began in September 2024 and will gradually be applied to all YouTube ads by 2026." The June 2026 update extended it to Search, stating that "Google may limit ad impressions from unqualified advertisers on searches that are more likely than others to result in negative ads experiences." August took it to everything, finishing in 2028.
Each step is recorded in Google’s 2026 policy change log as a routine entry. So: YouTube first, Search second, all Google Ads third, over four years. What is being built is an advertiser qualification regime across Google’s entire advertising business, introduced one surface at a time, in increments small enough that each one reads as a routine policy note.
"Unqualified" is a status you can hold without knowing why
Here is where the practical difficulty starts.
Google lists the factors it evaluates: "Account attributes," "User activity and reports," "Account maturity," "Ad format usage," and "History of policy compliance." That is a reasonable and honest list. What it is not is a specification.
There is no published threshold, no weighting, no score you can look up, and no stated ranking among the five. You cannot calculate your status, you cannot confirm you have cleared it, and you cannot tell which factor is holding you back. The policy says Google "will automatically review and update advertisers’ ad serving limits," so the status is dynamic, reassessed on a cadence Google does not publish.
The remedy on offer is an appeals form. Appealing requires knowing you have been limited, which brings us to the part that actually costs people money.
The failure mode is invisible, which is the real problem
A disapproved ad is a loud failure. It has a status, a reason, a help article and a fix. Every advertiser has dealt with one.
Limited ad serving produces a quiet failure instead. Your ads are eligible. Your campaigns are active. Your quality signals look normal. You simply appear less often, specifically in the scenarios most likely to generate complaints, and the impressions you lose are ones you never see. There is an in-account notification, which is genuinely useful and genuinely easy to miss in an interface full of them.
From inside the account, that presents as underperformance rather than restriction. It looks like a bidding problem, a seasonality dip, a competitor outspending you, a landing-page issue. It is none of those, and none of the usual advertising levers move it. Teams re-bid, rewrite ad copy, restructure campaigns and change agencies over symptoms that no amount of optimisation will resolve, because the constraint is not in the auction. Anyone tracking return on ad spend as their primary signal will see the number move and reach for the wrong lever.
The diagnostic tell is a drop in impression volume that is not matched by a corresponding change in your bids, budgets, quality metrics or competitive landscape, concentrated in some query types and not others.
Account maturity means new advertisers start behind
One named factor deserves separate attention because of what it structurally implies.
"Account maturity" is not a behaviour. It is a function of time. A new advertiser has no maturity by definition, no compliance history to speak of, and no accumulated user-activity signal. On the factors Google lists, a brand-new legitimate account and a brand-new abusive account look considerably more alike than either looks like a ten-year-old account in good standing.
That is defensible as anti-abuse design. Most ad fraud arrives through new accounts, and requiring accounts to earn full delivery is a reasonable way to raise the cost of that. It is also, unavoidably, a barrier to entry that falls hardest on exactly the advertisers least equipped to diagnose it: new businesses, first-time advertisers, and small teams without an agency who will read reduced impressions as a market signal about their product rather than a status on their account.
Established high-spend accounts will not notice this policy at all. That asymmetry is worth naming plainly, because it is the practical effect regardless of intent.
Brand clarity is an arbitrage crackdown in policy clothing
The June update named two criteria with unusual specificity: patterns of user complaints about an advertiser’s content or behaviour, and brand clarity, in the sense of preventing user confusion about who the advertiser actually is.
That second one is aimed at a recognisable set of practices. Ads that imply they are the brand they are reselling. Lead-generation pages that look like the manufacturer. Affiliate intermediaries that obscure who you are transacting with. Thin arbitrage sites that exist to monetise a click.
None of that is new, and Google has policies against the sharper versions already. What is new is that ambiguity about advertiser identity now feeds a delivery-limiting status rather than only a disapproval. You can be entirely within the letter of the content policies and still be judged unclear about who you are, which is a materially different compliance surface than the one most advertisers audit against. It is the same shift toward opaque, continuous evaluation we traced when AI citations decoupled from ranking: the rules stop being a checklist and start being a score.
What to do if your impressions dropped
Five positions, in the order I would take them.
Check your in-account notifications before you change anything. This is the only place Google tells you directly, and it is the step everyone skips because notifications are noise until one of them is not.
Look at impression volume, not cost or conversions. Limited ad serving suppresses appearances. If impressions fell without a matching change in budget, bid strategy, quality signals or competition, that pattern is the signal, and it will not show up cleanly in downstream metrics.
Audit brand clarity honestly. Does your ad, your display URL and your landing page make it obvious who the advertiser is, without a reader needing to infer it? If a reasonable person could think they are dealing with a brand you merely resell or represent, fix that first. It is the most actionable of the named criteria.
Treat account age as an asset you are building. Consolidating into one durable account with a clean compliance record is now worth more than it used to be, in the same way that consent and measurement setup became infrastructure rather than a checkbox. Spinning up fresh accounts to escape problems resets the one factor that only time can supply.
File the appeal if you are limited. The Limited Ad Serving Appeals Form exists, and an unappealed limit persists while automated review continues in the background. It costs little and is the only lever the policy explicitly offers.
The wider point is that the criteria for being allowed to advertise at all are becoming continuous, automated and unpublished, in the same way that the criteria for ranking did twenty years ago. That transition took search marketing a decade to absorb, and this one has until 2028.
Frequently Asked Questions
What is limited ad serving?
It is a Google Ads policy under which Google reduces how often an advertiser’s ads are shown, rather than disapproving them, when that advertiser is classed as unqualified. Google describes the goal as protecting the integrity of the Google Ads advertising ecosystem by limiting impressions of ads more likely to result in negative ads experiences. The key distinction is that the ads remain eligible and active; they simply appear less often in certain scenarios.
What changed in August 2026?
Google’s policy update, posted August 5, 2026, states that the policy expands to cover all Google Ads. It also states that implementation will begin gradually and will be completed by 2028, and that ads from unqualified advertisers will be limited in specific scenarios to decrease the potential for negative user experiences. Before this, enforcement had reached YouTube starting September 2024 and Google Search in June 2026.
How do I know if I am affected?
Google provides an in-account notification, which is the authoritative answer and worth checking first. Beyond that, the diagnostic pattern is a fall in impression volume that is not explained by changes to your budgets, bids, quality signals or competitive landscape, often concentrated in some query types rather than spread evenly across the account.
What makes an advertiser “qualified”?
Google names five factors it evaluates: account attributes, user activity and reports, account maturity, ad format usage, and history of policy compliance. It does not publish thresholds, weightings or a score, so there is no way to calculate your own status or confirm you have cleared the bar. Google states that it automatically reviews and updates advertisers’ ad serving limits, so the status is reassessed over time rather than fixed.
Is this the same as having ads disapproved?
No, and the difference matters for diagnosis. A disapproval is an explicit status with a stated reason attached to a specific ad, and it is visible in the interface. Limited ad serving leaves your ads approved and running while reducing how often they are served in particular scenarios. That is why it reads as underperformance rather than as a restriction, and why teams often spend months optimising against it.
Does this hit new advertisers harder?
Structurally, yes. Account maturity is one of the named factors, and a new account cannot have it, nor can it have an established compliance history or accumulated user-activity signal. That is a reasonable anti-abuse design, since much ad fraud originates in new accounts, but the practical effect is that new and small advertisers are more likely to be limited and least likely to have the expertise to recognise why.
Can I appeal?
Yes. Google provides a Limited Ad Serving Appeals Form. Because automated review continues regardless, a limit can also be lifted without an appeal if the underlying factors change. Appealing is worth doing anyway, since it is the only explicit remedy the policy offers and there is no cost to using it beyond the time it takes.
What does “brand clarity” mean here?
The June 2026 update highlights preventing user confusion about advertiser identity. In practice that points at ads, display URLs and landing pages that leave a reader unsure whether they are dealing with the brand itself, a reseller, an affiliate or an intermediary. You can comply fully with content policies and still be judged unclear on identity, which makes it a different audit than most advertisers currently run.
