Skip to content

How to fix: Unknown demographics excluded from campaign

finding google ads updated 2026.05.25 7 min read

How to fix: Unknown demographics excluded from campaign

TL;DR

The Unknown bucket in Google Ads demographic targeting holds every user whose age, gender, parental status, or household income Google could not confidently infer — cookieless sessions, privacy-restricted browsers, EEA traffic with denied consent, brand-new users with no signed-in profile. On most Search campaigns that bucket is 30-50% of otherwise-eligible impressions. Excluding it almost always shrinks volume without lifting conversion quality, because converters distribute across the known and Unknown buckets at similar rates. Remove the Unknown exclusion unless you have a hard compliance reason to keep it [1].

Why it matters

Google's demographic categories — age, gender, parental status, household income — are inferred from signed-in profile data, browsing behaviour, and aggregated signals. When Google cannot infer a value with confidence, the user lands in the Unknown bucket for that dimension [1]. The share of Unknown traffic has grown sharply since 2021: Safari Intelligent Tracking Prevention, Firefox Enhanced Tracking Protection, the deprecation of third-party cookies in Chrome, Consent Mode v2 in the EEA, and the steady rise of incognito and signed-out browsing all push real, high-intent users into Unknown.

The mistake usually starts in a reporting clean-up pass. A practitioner opens the Demographics report, sees an "Undetermined" or "Unknown" row taking 30-40% of impressions, decides it looks like noise, and excludes it. The exclusion drops impression volume by exactly that share, and the report now looks cleaner — but conversions drop in proportion to impressions, because Unknown is not a low-quality bucket. It is a missing-signal bucket. The user behind that Unknown impression may be a brand-new Safari user signed out of their Google account, who is also your highest-converting persona.

Google's own documentation explicitly warns against this exclusion: "Exclude the 'Unknown' demographic category only if you're sure you want to restrict your campaign to a narrow audience. Excluding 'Unknown' might prevent a substantial number of people from seeing your ads, some of whom you might want to reach." [1] The default for every new campaign is to include Unknown precisely because the Unknown bucket overlaps the converting audience.

There are two legitimate exceptions. Age-gated verticals (alcohol, gambling, adult products) must exclude Unknown age to comply with platform policy and local law — they cannot serve to "could be under 18." Income-gated products (private wealth, luxury concierge) sometimes exclude Unknown household income to keep budget on high-bracket users. Both are compliance moves, not optimization moves, and both should be documented in the account notes so a future auditor does not undo them.

For everyone else, an Unknown exclusion is a self-inflicted reach cap. Smart Bidding can no longer optimize across the full inventory; the auction sees fewer queries; CPC tends to rise as the algorithm bids harder on the remaining smaller pool to hit volume targets.

How to fix

  1. Open the Demographics view for the affected campaign (Campaign → Audiences, keywords and content → Demographics). Inspect each dimension tab: Age, Gender, Parental status, Household income.
  2. Locate the Excluded list at the bottom of each tab. Look for rows labelled Undetermined age, Unknown gender, Parental status undetermined, or Household income undetermined. Each one is a separate dimension — Unknown gender and Unknown age are excluded independently, and an account can have only one or two excluded.
  3. Remove the Unknown exclusion on each dimension by hovering the row and clicking the remove icon. If the campaign sits in an age-gated vertical (alcohol, gambling) or income-gated vertical (regulated finance, luxury), leave the relevant Unknown exclusion in place and document why in the account notes.
  4. For Manual CPC campaigns that want softer treatment of Unknown traffic, switch from exclusion to a small negative bid adjustment (-10% to -20%) on the Unknown row. This keeps the inventory in the auction at a slightly lower bid rather than zeroing it out. Do NOT use this approach on Smart Bidding campaigns — tCPA, tROAS, Maximize Conversions, and Maximize Conversion Value ignore manual demographic bid adjustments [2], so the modifier shows in the UI but has no auction effect.
  5. Repeat the audit for every Search campaign in the account. Demographic exclusions are configured at campaign level, so each campaign needs the same pass. A blanket account-level Unknown exclusion does not exist, but practitioners often apply the same exclusion across campaigns via mass edits.
  6. Verify after 14 days in the Demographics report. Unknown rows should now show impressions and conversions. Overall campaign impression share should rise. Conversion rate should hold within roughly 10% of baseline; if it drops further, the Unknown bucket on that campaign genuinely was low-quality (rare) and a soft bid adjustment is the better posture.

Common mistakes

  • Excluding Unknown to clean up reports. This is the most common origin. The report looks tidier but the campaign shrinks by 30-50% of impressions, and conversions shrink with it. Reporting clarity is not worth losing converters.
  • Assuming Unknown means "low quality" or "fraudulent." Unknown is a missing-signal label, not a quality label. The bucket includes Safari users, EEA users with denied consent, signed-out Chrome users, and brand-new visitors — many of whom are high-intent converters.
  • Excluding Unknown household income on a B2C product with a broad price point. Household income inference is sparse to begin with and skewed toward signed-in Google account users with sufficient signal. Excluding Unknown household income on a consumer product typically removes 40-60% of impressions because Google rarely has high-confidence income data outside the US.
  • Adding demographic bid adjustments on Smart Bidding campaigns. Bid modifiers on tCPA / tROAS / Maximize Conversions / Maximize Conversion Value are ignored at auction time [2]. Use targeting (include/exclude) decisions on Smart Bidding, not bid adjustments. On Manual CPC, modifiers still apply.
  • Forgetting to undo the exclusion after a vertical change. Accounts that pivoted from a regulated vertical (gambling, alcohol) to a general consumer product sometimes carry the age-gated Unknown exclusion forward. Audit demographic exclusions whenever campaign category or product fit changes.

FAQ

What share of my traffic actually lands in Unknown? On Search campaigns in 2025, expect 25-40% Unknown on gender and age, 50-70% Unknown on parental status, and 60-80% Unknown on household income (income inference is the sparsest). EEA accounts with Consent Mode v2 enforced see higher Unknown shares than rest-of-world accounts because denied consent maps to Unknown.

Will removing the Unknown exclusion hurt my conversion rate? Usually not by more than 5-10%, because converters distribute across known and Unknown buckets similarly. If conversion rate drops sharply after removal, the Unknown bucket on that campaign was genuinely lower-quality (rare) and a soft -15% bid adjustment on Manual CPC, or no adjustment on Smart Bidding, is the better posture.

Does this apply to PMax and Demand Gen campaigns? Demographic exclusions exist on PMax (campaign-level demographic exclusions added in 2024) and on Demand Gen, but the audit lens is the same: do not exclude Unknown unless compliance requires it. PMax's audience expansion will still surface the right users when signals allow.

What about Display and YouTube campaigns? Same principle. Unknown shares on YouTube tend to be higher because more viewers are signed out or on cookieless devices. Excluding Unknown on Display or YouTube typically removes 40-60% of impressions.

Is there a legitimate use of the Unknown exclusion? Yes — age-gated and income-gated verticals where serving to "could be under 18" or "could be low-income" violates platform policy or product positioning. Document the reason in account notes so a future auditor does not undo the exclusion.

Sources

  1. Google Ads Help — About demographic targeting. Defines the Unknown demographic category as users whose age, gender, parental status, or household income Google could not identify, and warns that excluding Unknown "might prevent a substantial number of people from seeing your ads, some of whom you might want to reach."
  2. Google Ads Help — About Smart Bidding. Documents that Smart Bidding factors demographic signals automatically at auction time and ignores manual demographic bid adjustments on tCPA / tROAS / Maximize Conversions / Maximize Conversion Value campaigns.
// was this useful?
// anonymous · no personal data stored