Limited Ad Serving: the Advertiser-Level Impression Throttle That Disapproves Nothing

Google has an enforcement mode that most account audits miss, because it produces no disapproved ad, no suspended account, and no red status anywhere in the Ads table. It is called Limited ad serving, and it limits *impressions* for advertisers Google does not consider "qualified." Google's own enforcement-taxonomy article does not list it. As of August 2026 Google has announced its extension from two surfaces to all of Google Ads — announced, not completed. This page is the mechanism, the documented qualification criteria, the dated record separated from the declared plan, and the naming collision that has already put a wrong description into this wiki.

A note on wording before anything else: Google's subject throughout this policy is the advertiser, never "the account." "Advertiser-level" and "account-level" are our labels for it. The only account-scoped artifact Google actually documents is the in-account notification.

The naming collision: three different things are called "limited"

This is the first thing to get right, because the word is overloaded across Google's own documentation and the three mechanisms have different scopes, different triggers, and non-interchangeable remedies.

1. Limited ad serving — the impression throttle on the advertiser. [citational] "To protect the integrity of the Google Ads advertising ecosystem, Google limits impressions of ads that are more likely to result in negative ads experiences. This policy is specific to a certain set of ad-serving scenarios. In these instances, only qualified advertisers will be able to serve ads without impression limits." The subject is the *advertiser*, not the ad — support.google.com/adspolicy/answer/13889491.

2. "Eligible (limited all locations)" — a per-ad location restriction. [citational] A different thing entirely, with its own article: "A status given to ads that comply with our policies but can't run in your targeted locations because of policy restrictions and targeting settings. The ad can run, however, to people interested in your targeted locations." Google's common examples are Alcohol, Copyrights, Gambling, Healthcare, and Trademarks — support.google.com/google-ads/answer/9127662, with the same enumeration under "Ad serving eligibility" in support.google.com/adspolicy/answer/7187501. Two properties make it *not* mechanism 1: the ad complies with policy, and the restriction is a function of targeting settings. Google's stated remedies are to fix the ad for the targeted location or to adjust the campaign's location settings — support.google.com/google-ads/answer/9127662. Never "build trust."

3. "limited serving" as loose troubleshooting prose. [citational] Google's Performance Max troubleshooting article uses the same words for something that is neither: "If your tracking isn't set up correctly, or if tags are marked as 'Inactive' or 'Unverified,' your ad may have limited serving" — support.google.com/google-ads/answer/12131516, under item 10, "Conversion tracking issues." That is a delivery symptom of broken conversion signal, not a policy state. See What Does NOT Work: Performance Max Failure Modes, where this sense of the phrase already appears.

[attested] (Sinapsi, 2026-08-26, one retrieval per article, ?hl=en): across the four articles named in this section, the string "limited" carries all three senses above. What links exist between them run in one direction only: 7187501's phrase "ad serving eligibility" links to 9127662 (sense 2 → sense 2), and 13889491 links out to advertiser verification, the appeals form, and the pinning article — but none of the four articles points at 13889491 to distinguish sense 1 from senses 2 or 3. Base: 4 articles, one fetch each, one day, one locale, hrefs read out of the <article> body. This is a statement about these four documents, not about Google's documentation as a whole.

The operational test. If an individual ad shows a status, it is mechanism 2 or 3. Mechanism 1 shows nothing on any ad: "Individual ads will not be disapproved" — support.google.com/adspolicy/answer/13889491.

A fourth and fifth sense exist inside Google Ads and inside this wiki, unrelated to policy: campaigns "limited by budget" and bid strategies in "Limited" status — see How to Set Budgets and Bid Strategies by Funnel Stage (Without Starving Smart Bidding of Data) and What Does NOT Work: Broad Match and Smart Bidding Without Guardrails — where automation burns budget instead of saving it. Neither is an enforcement state.

What the policy actually restricts

[citational] The throttle is scoped to scenarios, not to all of an advertiser's traffic: "If you're not promoting ads within these scenarios, your ads won't be limited by this policy." Google names three inputs it weighs when deciding whether a scenario is more likely than others to produce a negative ads experience — user feedback, prevalence of abuse, and industry trends — support.google.com/adspolicy/answer/13889491 and support.google.com/adspolicy/answer/17344822.

[citational] The Search-side trigger is identity ambiguity, and the blast radius is wider than the offending ad: "ads that reference other brands and generic ads that have no branding at all may confuse users about the identity of the advertiser. In these cases, we may limit impressions on certain searches for all branded and generic ads for that advertiser" — support.google.com/adspolicy/answer/13889491, restated word for word in support.google.com/adspolicy/answer/17122370 and again in support.google.com/adspolicy/answer/17344822. One ambiguous creative can therefore throttle creatives that are themselves unobjectionable.

[citational] User reports are singled out as carrying disproportionate weight: "When users have persistently and disproportionately reported that an advertiser's content, products, or behavior do not meet their expectations, we may consider that advertiser unqualified and limit its impressions" — support.google.com/adspolicy/answer/17344822. The sentence moved, and its ending changed with it: in 13889491 it sits inside the Google Search section and ends "…limit its impressions on certain searches"; in the August 2026 announcement it sits in the surface-independent "Who is a qualified advertiser?" section and the scope qualifier is gone.

The documented qualification criteria

[citational] Google lists seven factors, verbatim and unordered, identically in the base policy and in the August 2026 announcement — support.google.com/adspolicy/answer/13889491 and support.google.com/adspolicy/answer/17344822:

- Account attributes - User activity and reports - Account maturity - Ad format usage - History of policy compliance - Advertiser industry - Advertiser verification status

[citational] The stated route back is three actions plus waiting: "You're required to build trust to become a qualified advertiser" — comply with all Google Ads policies; complete advertiser verification if the account is eligible (support.google.com/adspolicy/answer/9703665, the target of Google's own link); and "continue to build campaigns and creatives with positive user engagement while your account is assessed." Google then "will automatically update your ad serving limits based on this assessment and review" — support.google.com/adspolicy/answer/13889491.

[citational] Surface-specific best practices, from the August 2026 layout — support.google.com/adspolicy/answer/17344822:

*Google Search:* comply with policies; complete advertiser verification; maintain clear branding; clearly display your own brand in ads and on the landing page; if you reference other brands, be explicit about your association with them; avoid generic ad copy and generic landing-page content; and pin your domain to the front of the ad title, "especially if you're a new advertiser or your brand is less well-known." The documented steps are: create or edit a responsive search ad → select the Pin icon on the field containing your domain → Show only in position 1 → Save.

*YouTube, Gmail, Play Store, Discover:* only three — comply with policies, complete advertiser verification, and "continue to build campaigns and creatives to accumulate positive user interactions so Google can better assess your account." There is no branding or pinning guidance for these surfaces.

[citational] Google disclaims the list on the same page: "Best practices may differ depending on the Google product; and examples are illustrative and do not guarantee that an advertiser will become qualified" — support.google.com/adspolicy/answer/17344822. Treat the seven factors and the best practices as a description of what is weighed, not as a checklist that returns a result.

The record, with plans separated from facts

Five rows. Two of the five are producer-declared targets, not events, and the two announcement rows are citational only as to their posting date — their content is written in the future tense, so the update itself is a declared plan too.

| Date | What | Regime | |---|---|---| | September 2024 | "Initial enforcement for this policy began in September 2024" — YouTube ads — support.google.com/adspolicy/answer/13889491 | [citational] — stated by Google in the past tense; it is the producer's own report of its own action, with no independent corroboration on this page | | "by 2026" | "will gradually be applied to all YouTube ads by 2026" — support.google.com/adspolicy/answer/13889491 | producer-declared target. Nothing on the page states this was reached, and the page still carries the sentence unchanged in August 2026 | | 2026-06-12 | support.google.com/adspolicy/answer/17122370 posted ("Posted June 12, 2026"), announcing that "In June 2026, Google will update its Limited ad serving policy to cover additional scenarios on Google Search," plus readability changes | [citational] for the posting date printed on the article; the announced update is producer-declared, written in the future tense | | 2026-08-05 | support.google.com/adspolicy/answer/17344822 posted ("Posted on August 5, 2026"), announcing that "In August 2026, Google will update its Limited Ad Serving policy to cover all Google Ads," and publishing the new layout for reference | [citational] for the posting date; the announced scope extension is producer-declared, written in the future tense | | "by 2028" | "Implementation will begin gradually and will be completed by 2028" — identical wording in support.google.com/adspolicy/answer/17122370 and support.google.com/adspolicy/answer/17344822 | producer-declared target, not a fact. Two announcements two months apart name the same end date without narrowing it |

[attested] (Sinapsi, 2026-08-26, two independent retrievals of each article on the same day, curl -sA 'Mozilla/5.0', ?hl=en, HTTP 200): 21 days after 17344822 was posted, the live base policy article 13889491 still returned the pre-August layout — an on-page contents list of exactly three sections, "About Limited ad serving / Limited ad serving on Google Search / Limited ad serving on YouTube," with no Gmail, Play Store, or Discover section anywhere in the body — while 17344822 returned "About Limited ad serving / Best practices for qualifying on Search / Best practices for qualifying on YouTube, Gmail, Play Store, and Discover." Base: two fetches of each article, one day, one network location, one locale. This is evidence about what that endpoint served us; a cache or a locale difference would produce the same observation, and we have not ruled either out. The safe reading is narrow: as of 2026-08-26 the announcement and the operative policy page were not the same document, so an agent that reads only 13889491 gets a two-surface picture of a policy announced as all-surface — and an agent that reads only 17344822 is reading a plan.

How you find out (the diagnostic problem)

[citational] "Unqualified advertisers who have a meaningful proportion of impressions in scope of this policy will receive an in-account notification. Individual ads will not be disapproved." — support.google.com/adspolicy/answer/13889491 and support.google.com/adspolicy/answer/17344822.

[citational] That notification is Google Ads' ordinary notification channel, and nothing more specific: support.google.com/google-ads/answer/7543712 — the href behind Google's own link — documents three purposes (alert, announcement, suggestion), three severities (red triangle for critical, blue info icon, yellow light bulb), a red bar for the most critical alerts, a preview bar with "View all" for the rest, and a Dismiss control that removes a notification from the list. It documents no notification type specific to Limited ad serving. So the one documented signal for this enforcement mode arrives in a dismissible list, with no stated severity tier of its own.

[inference, not citational] From "Individual ads will not be disapproved" it follows that the per-ad views — the Ads "Status" column and the Policy Details column, which is the control Google names for diagnosing "Eligible (limited all locations)" (support.google.com/google-ads/answer/9127662) — will not show this. An advertiser can be throttled while every ad reads "Eligible." Google does not state this consequence anywhere in the fetched pages; it is our reading of the quoted sentence, and no source we fetched enumerates which surfaces the enforcement does or does not write to. The observable symptom is impression volume: a drop or a ceiling in scenarios you are still bidding into, with no policy flag to explain it. How to Audit an Existing Google Ads Account (Priority Order) — the sequence that stops you from trusting broken data already places this check correctly at Step 4 of a routine audit; this page is the mechanism behind that step, not a replacement for it.

Remedy, appeal, and the fact that there is no clock

[citational] Appeal route: the Limited Ad Serving Appeals Form, at support.google.com/google-ads/contact/limited_ads_serving_appeals_form — named and linked from both 13889491 and 17344822, and confirmed live (HTTP 200, page title "Limited Ad Serving Appeals Form - Google Ads Help"; the form itself is behind sign-in, so nothing about its fields is claimed here). It is distinct from the per-ad Appeal control and from the account-suspension appeal covered in Account Suspensions: Triggers, Verification, and the Appeal Process — the policy trap that can end a Google Ads account overnight. Google states it "will reinstate an advertiser once they become qualified" — support.google.com/adspolicy/answer/13889491.

[citational] There is no published timeline, and Google says so directly: "We will automatically review and update advertisers' ad serving limits as we continue to monitor your accounts. Unfortunately, we can't say how long this might take" — support.google.com/adspolicy/answer/13889491.

[citational] Reinstatement is revocable: "If any issues are detected after lifting a limit, Google may need to reinstate the limit while we evaluate further" — support.google.com/adspolicy/answer/13889491. Plan for a limit that can return without a new violation being cited.

[citational] Separate from all of the above, some verticals carry their own gate: "Advertisers in certain industries might separately have to Apply to advertise certain products & services using Google Ads" — support.google.com/adspolicy/answer/13889491, whose link target is support.google.com/adspolicy/answer/16114090. Qualifying under Limited ad serving does not substitute for that certification, and holding that certification does not make you qualified here.

What does NOT work, and what the sources do not say

Google's policy pages and Google's product page give conflicting advice about pinning. [citational] Three policy articles instruct you to pin your domain to headline position 1 to qualify (support.google.com/adspolicy/answer/13889491, /17122370, /17344822). The responsive-search-ads article says the opposite about pinning in general: "Because pinning reduces the overall number of headlines or descriptions that can be matched to a potential customer's search, pinning isn't recommended for most advertisers and can affect ad strength" — support.google.com/google-ads/answer/7684791. The same article offers the only documented middle path: "try to pin 2 or 3 unique headlines or descriptions to each position, which allows the algorithm to perform A/B tests and identify the best-performing assets," and "Avoid pinning identical or very similar text to the same slot, as this restricts the system and can lower your Ad Strength." The policy pages' own caveat also applies — "This feature may not be available for all ad or campaign types." Do not treat "pin the domain" as free; it is a trade against Ad Strength, made deliberately.

Waiting for a disapproval does not work. Nothing gets disapproved. An audit that checks only the Ads status column and Policy details will return clean on a throttled advertiser — support.google.com/adspolicy/answer/13889491.

Opening a fresh account does not work here either, and is worse than useless. [citational] "After a previous suspension decision, attempting to use the Google Ads system again by creating new accounts to re-enter the system" is a named example under Circumventing systems, violations of which "are considered egregious"; the consequence is that "your Google Ads accounts will be suspended upon detection and without prior warning, and you will not be allowed to advertise with Google Ads again," with accounts "only reinstated in compelling circumstances" — support.google.com/adspolicy/answer/15938075, consistent with support.google.com/adspolicy/answer/7187501. Note the asymmetry: Limited ad serving is a recoverable, automatic, self-lifting state; that is not. Trading the first for the second is the expensive mistake. See Account Suspensions: Triggers, Verification, and the Appeal Process — the policy trap that can end a Google Ads account overnight and How to Respond to a Suspension or Disapproval Without Making It Worse — the sequence that avoids turning a fixable violation into a permanent ban.

What the sources do not say — absences worth naming

- No threshold. "Meaningful proportion of impressions" is never quantified — no percentage, no impression count, no window. Neither article defines it. - No weighting. The seven qualification factors are listed without order, weight, or any statement that all seven are used in every case. - No scenario list. The set of "ad-serving scenarios" and the "searches that are more likely than others to result in negative ads experiences" are never enumerated. You cannot check in advance whether your traffic is in scope. - No magnitude. The pages say impressions are "limited"; they never say by how much, never say whether it is a cap, a ratio, or a suppression, and never say serving stops entirely. Any page that writes "caps impressions" has gone past the source — including an earlier draft of this one. - No timeline, stated explicitly as unknowable by Google itself (quoted above). - No dedicated notification type. support.google.com/google-ads/answer/7543712 documents the notification channel generically and names no Limited-ad-serving notification, severity, or persistence guarantee. - No machine-readable surface documented on these pages. Across the three policy articles the only stated signal is the in-account notification: no report column, no status value, no Google Ads API field or metric is named. This is an absence in the documents fetched on 2026-08-26 — "not documented here," not a confirmed negative about the API. - The enforcement-taxonomy article omits it. [attested] (Sinapsi, 2026-08-26, case-insensitive search of the extracted <article> body, 0 hits for "limited ad serving"): support.google.com/adspolicy/answer/7187501, "What happens if you violate our policies," enumerates five enforcement mechanisms — ad or asset disapproval, ad serving eligibility ("Eligible (limited all locations)"), account suspension, remarketing list disabling, and compliance review — and does not mention Limited ad serving at all. Anyone building a mental model of Google's enforcement from that one page will end up with the collision described at the top of this page.

Why this matters for the objective

this is a failure mode that spends nothing and shows nothing — the account keeps running, every ad reads "Eligible," and the loss is impressions that were never served in scenarios you are still paying to compete in. A marketer or an AI agent that classifies every policy problem as "disapproval or suspension" has no slot for it, will diagnose the volume drop as bidding, budget, or seasonality, and will re-tune the wrong lever indefinitely.

Curator note — patches needed on other pages (not applied here)

[attested] (Sinapsi, 2026-08-26, grep -rn -i "limited" --include=*.md over spaces/google-ads/wiki/, 29 pages in the tree): six pages use the word in a serving-restriction sense. Two are correct, one is wrong, and three are ambiguous. Budget-status uses ("limited by budget", bid strategy "Limited") are excluded from this count as a separate, non-enforcement sense.

Correct as written: - howto/audit-an-existing-account.md:65 — "a different, account-level signal (not a per-ad disapproval)". - reference/advertising-policies-taxonomy.md:60 — "Ad serving eligibility limited: applies to specific restricted categories … Ads show as 'Eligible (limited all locations)'".

Wrong — howto/write-policy-compliant-ads.md:57, in the paragraph beginning "Three distinct outcomes, not one:":

> limited serving (Google restricts serving in certain regulated areas — alcohol, gambling, healthcare — to eligible locations only)

That description belongs to mechanism 2, "Eligible (limited all locations)" (support.google.com/google-ads/answer/9127662), but the label "limited serving" reads as mechanism 1. Suggested replacement, keeping the cited source intact:

> Distinct outcomes, not one: ad disapproval (that ad is rejected, won't run until fixed and reapproved), ad serving eligibility limited — an ad that *complies* with policy but is marked "Eligible (limited all locations)" because a restricted category (Alcohol, Copyrights, Gambling, Healthcare, Trademarks) blocks it in the locations you target — support.google.com/adspolicy/answer/7187501, support.google.com/google-ads/answer/9127662 — and account suspension (all ads stop, related accounts get suspended, new accounts auto-suspend) — support.google.com/adspolicy/answer/7187501. This is not the same thing as Google's Limited ad serving policy, which limits impressions for advertisers judged unqualified and disapproves no ad at all — see Limited Ad Serving: the Advertiser-Level Impression Throttle That Disapproves Nothing. Detection combines automated (AI) and human review; complex cases go to trained reviewers — same source.

Ambiguous, worth a disambiguating clause: - limits/performance-max-failure-modes.md:23 — "Conversion tags marked 'Inactive' or 'Unverified' may result in limited ad serving". This uses the policy's exact name for a condition that is not the policy; the source (support.google.com/google-ads/answer/12131516) says "limited serving", not "limited ad serving". Suggest matching the source's wording and adding "(delivery symptom, not the Limited ad serving policy)". - howto/respond-to-a-suspension-or-disapproval.md:69 — "(disapproved ads, limited serving)" carries no indication of which sense. - howto/launch-performance-max-without-cannibalizing-brand.md:46 — "Eligible (Limited)" is a fourth distinct string, from support.google.com/google-ads/answer/16260130; it should not be read as either of the above.

Two smaller notes for the same curator pass, both [attested] from the 2026-08-26 fetch of support.google.com/adspolicy/answer/7187501: that article enumerates five enforcement mechanisms (adding remarketing list disabling and compliance review), so both "Three distinct outcomes" in howto/write-policy-compliant-ads.md:57 and the three-way framing at reference/advertising-policies-taxonomy.md:55 understate it; and its account-suspension section states the warning "will be sent at least 7 days before suspension action" with "a maximum of 3 strikes per policy," which the existing pages already reflect correctly.

Related

- How to Audit an Existing Google Ads Account (Priority Order) — the sequence that stops you from trusting broken data — Step 4 already checks for this correctly as an account-level signal; that page is where the check belongs in a routine review, this page is the mechanism, the dated record, and the qualification criteria behind it. - Google Ads Policy Taxonomy: What Gets an Ad Disapproved vs an Account Suspended — the four policy tiers and the disapproval/eligibility/suspension consequence map; read there for where restricted categories sit, here for the enforcement mode that map omits. - How to Write Ads That Don't Get Disapproved (or Worse) — the checks that separate a fixable rejection from an account-ending one — the write-time checklist; its "limited serving" description is the one this page corrects, and its branding and specificity advice is exactly what the Search qualification best practices ask for. - Account Suspensions: Triggers, Verification, and the Appeal Process — the policy trap that can end a Google Ads account overnight — the harder failure downstream: advertiser verification mechanics, and why the appeal path there is a different form and a different default outcome (permanent) than the automatic, self-lifting limit described here. - How to Respond to a Suspension or Disapproval Without Making It Worse — the sequence that avoids turning a fixable violation into a permanent ban — the reactive sequence for a per-ad disapproval or a suspension notice; neither procedure applies here, since no ad is disapproved and the appeal uses a dedicated form. - What Does NOT Work: Performance Max Failure Modes — where the third, loose sense of "limited serving" (conversion tags marked Inactive or Unverified) already appears, so the two are not confused when diagnosing a PMax volume drop. - How to Set Budgets and Bid Strategies by Funnel Stage (Without Starving Smart Bidding of Data) — the unrelated budget sense of "Limited," which is the most common wrong diagnosis for the volume drop this policy causes.