DEV Community

Alexander Todosuik
Alexander Todosuik

Posted on

AI Max Quality Score: How Ad Relevance, Landing Page Experience, and Expected CTR Are Calculated

Quality Score in Google Ads measures ad quality on a 1-10 scale across three components: Expected Click-Through Rate (CTR), Ad Relevance, and Landing Page Experience. For traditional Search campaigns, Quality Score directly affects ad rank and cost-per-click. In AI Max campaigns, Quality Score works differently — but it still matters.

How Quality Score Works in AI Max

AI Max doesn't use traditional keyword-based Quality Score the way Search campaigns do. There are no keywords to calculate expected CTR for; AI Max's query matching is dynamic. However, Quality Score-equivalent signals still influence AI Max performance:

  • Ad Relevance signals inform how well AI Max can match assets to queries
  • Landing Page Experience directly affects the campaign's effective quality
  • Expected CTR is modeled at the asset level, not keyword level

In Google Ads, AI Max campaigns show Quality Score data in the "Asset groups" view and in the "Assets" performance view, not in a keyword-level report. Asset performance ratings ("Best," "Good," "Low") are partly informed by Quality Score factors.

The practical implication: you won't see a "7/10 Quality Score" for AI Max the way you do for keywords. But improving the three underlying components — relevance, landing page quality, and expected CTR — directly improves AI Max performance even if the number isn't explicitly surfaced.

Expected CTR in AI Max

For AI Max, expected CTR is calculated at the asset level. An asset that consistently gets high CTR in contexts where it's shown will score "Best" in asset performance ratings. An asset with low CTR scores "Low" and gets fewer impressions over time.

How to improve expected CTR:

  • Write headlines that match search intent explicitly (specificity improves relevance)
  • Use numbers and specifics: "Get 3 quotes in 5 minutes" vs. "Get quotes fast"
  • Ask a question relevant to the searcher's intent: "Need a faster payroll solution?"
  • Match the headline to the search context AI Max will deploy it in

Review your "Low" rated assets in the asset performance view and identify common patterns. Low CTR assets often:

  • Use generic value propositions that don't differentiate
  • Lead with company name instead of benefit
  • Make claims that are obviously true for all competitors too

Ad Relevance in AI Max

Ad Relevance measures how closely your ad content matches the search query. For AI Max, relevance is computed across the entire asset group, not individual keyword-ad pairs.

Factors affecting AI Max ad relevance:

  • Asset group headline vocabulary match to actual search queries that trigger the campaign
  • Description content alignment with what users are searching for
  • Final URL landing page content alignment with asset group messaging

If your AI Max campaign is showing for irrelevant queries (visible in search terms report), relevance scores are lower and costs are higher. The fix is usually a combination of:

  1. Adding negative keywords to block irrelevant queries
  2. Adding more targeted headlines that better describe your specific offering
  3. Improving landing page content alignment

Structural relevance: Each asset group should focus on one theme. A single asset group mixing "accounting software" and "legal software" has poor relevance for either query type. Separate asset groups by theme improve relevance for each.

Landing Page Experience

Landing Page Experience measures:

  • Page load speed (Core Web Vitals: LCP, FID/INP, CLS)
  • Mobile friendliness
  • Content relevance to the ad
  • User experience signals (bounce rate, time on site, pogo-sticking back to search)
  • Transparent and trustworthy content

Google measures these through Googlebot crawls, Core Web Vitals data from Chrome UX Report, and behavioral signals from search traffic.

Landing Page Experience checklist for AI Max:

Speed metrics (check with PageSpeed Insights):

  • LCP (Largest Contentful Paint) under 2.5 seconds
  • INP (Interaction to Next Paint) under 200ms
  • CLS (Cumulative Layout Shift) under 0.1

Mobile experience:

  • Viewport meta tag configured
  • No horizontal scrolling required
  • Touch targets (buttons, links) at least 44x44px
  • No intrusive interstitials that cover content on mobile

Content relevance:

  • Landing page headline matches ad headline (or closely related)
  • Primary CTA visible above the fold on mobile
  • No bait-and-switch content (ad promises X; landing page delivers Y)

Trustworthiness:

  • HTTPS (required)
  • Privacy policy accessible
  • Contact information visible
  • Real business signals (address, phone, about page)

The July 2026 ToS and Quality Score Impact

The July 2026 ToS update (https://yositeup.com/blog/google-ads-tos-july-2026-ai-automation-what-changed) included provisions that directly affect Quality Score components:

  • AI-generated headlines that are "low effort" or generic may be flagged and receive lower Ad Relevance ratings
  • Landing pages with excessive AI-generated thin content receive lower Landing Page Experience scores
  • Claims in ads that don't match landing page claims affect both Relevance and Landing Page Experience scores

For AI Max specifically: review all AI-suggested assets before enabling them. Google's automated asset suggestions may generate technically policy-compliant content that still receives poor performance signals because it's too generic to achieve high expected CTR.

Quality-Related Metrics to Monitor Weekly

In Google Ads for your AI Max campaigns:

Asset performance distribution: Go to Campaigns > Asset groups > Assets tab. Count "Best," "Good," and "Low" assets. Target: >50% "Best" or "Good," <20% "Low." If you have many "Low" assets, replace them with new variations.

Search impression share: If impression share is declining without budget reduction, quality factors are reducing your competitiveness. Check which quality components are degrading.

Landing page performance: In Google Analytics 4, filter sessions by "Google Ads / CPC" source and check landing page bounce rate and engagement. If specific landing pages have dramatically higher bounce rates, they need content or UX improvement.

Core Web Vitals: Google Search Console > Core Web Vitals. URL-level data shows which pages pass or fail. Failed pages serve AI Max traffic at a disadvantage.

Comparing Quality Across Multiple Asset Groups

For AI Max campaigns with multiple asset groups, compare:

  • Asset performance ratings by asset group
  • Conversion rate by asset group (Conversions / Clicks)
  • CPA by asset group

If one asset group consistently outperforms others on all metrics, examine what's different: is it the landing page, the headline themes, the audience signals? Replicate the success pattern in other asset groups.

The June 2026 reporting deletion (https://yositeup.com/blog/google-ads-reporting-data-deleted-june-2026) may have disrupted asset performance rating calculations for that period. Asset ratings update continuously based on recent performance; if a significant data gap occurred, some assets may have been rated on insufficient data. Post-July 2026 asset ratings are more reliable than those from the affected period.

The DSA to AI Max migration (https://yositeup.com/blog/google-ads-dsa-ai-max-migration-february-2027) improves quality score opportunities significantly: DSA's auto-generated headlines often scored poorly on Ad Relevance because they were extracted from page content rather than written for ad context. AI Max with manually crafted, test-and-iterate asset groups typically achieves better quality signals over time.

AI Max Shopping quality (https://yositeup.com/blog/google-ai-max-shopping-replacing-performance-max-2026) is partly determined by Merchant Center product data quality — accurate titles, complete attributes, high-resolution images — rather than headline-level quality scoring. The quality framework differs but the underlying principle is the same: relevance of content to searcher intent.

Top comments (0)