How to fix: Quality Score components rated Below Average

TL;DR

Quality Score is built from three diagnostic components — Expected CTR, Ad Relevance, and Landing Page Experience — and Google rates each one Above Average, Average, or Below Average per keyword [1]. When at least 30% of your enabled keywords carry a Below Average rating on any one of those three components, every click in that cluster pays an Ad Rank penalty in inflated CPC and reduced auction eligibility. The fix is component-specific: identify which of the three is rated Below Average per keyword, then apply the corresponding remediation — query hygiene for Expected CTR, ad group splits plus RSA rewrites for Ad Relevance, and landing page parity work for Landing Page Experience.

For the broader keyword-level Quality Score frame (median QS, share at QS ≤ 4, the case for fixing QS at all), see the keyword-level Quality Score fix article. This article is the component-by-component diagnostic that sits one level deeper.

Why it matters

Quality Score is the keyword-level multiplier in the Ad Rank formula. Google states directly that "your Ad Rank is calculated using your bid amount, auction-time ad quality... and other factors" [2], and the three diagnostic components are the levers ad quality is built from [1].

The mechanism: a single keyword with a Below Average rating on any one component drags overall Quality Score down even if the other two read Average or Above Average. Keywords with QS ≤ 4 lose auction eligibility against equally-bid competitors at QS 7+, and the clicks that do fire pay an inflated CPC to compensate for the relevance penalty. At ≥30% of enabled keywords carrying at least one Below Average component, the cost taxation compounds across the keyword set, undermining downstream optimisation work like Smart Bidding learning [3].

Critically, the three components are independent. A copy-only fix on a landing-page-rooted issue will not move Quality Score — you must read the component diagnostic columns before you start fixing. This is why the rule surfaces the per-component share rather than only the aggregate QS distribution.

This rule pairs with three sibling fixes that own specific component-level deep dives — Ad Relevance via RSA Ad Strength for the ad copy lever, and Landing Page Experience for the page-side lever. Read this article first to identify which lever to pull; read the sibling articles for the tactical playbook on that lever.

How to fix

Total time: 1-2 hours of diagnostic work, then 2-4 hours of remediation per failing component, plus a 21-day measurement window before component ratings refresh.

1. Surface the component diagnostic in the Google Ads UI

Open Search keywords (Campaigns → Audiences, keywords and content → Search keywords). Click Columns → Modify columns → expand the Quality Score group and enable all four columns:

• Quality score
• Exp. CTR
• Ad relevance
• Landing page exper.

Save the column set as "QS diagnostic" so you can recall it for future audits. Filter Status = Enabled and sort by Quality score ascending.

2. Group keywords by which component is Below Average

Export the keyword table and sort by the three component columns. Each enabled keyword falls into one of these diagnostic buckets:

Do not start fixing until you have classified each failing keyword by component — the wrong lever wastes the 21-day measurement cycle.

3. Fix Expected CTR Below Average — query hygiene

Expected CTR rates how likely your ad is to be clicked when shown for that keyword, relative to the keyword's ad position [4]. Below Average means Google sees too many irrelevant impressions for that keyword.

• Export 90 days of search terms (Campaigns → Insights and reports → Search terms) and identify queries with high impressions and zero conversions.
• Add 50-200 tightly-targeted negative keywords sourced from that export. Use exact-match negatives for problem queries and phrase-match negatives for problem patterns.
• Pause structural-loss keywords: QS ≤ 3 with ≥1,000 impressions in 90 days and zero conversions are unlikely to recover. Pausing them lets the remaining set re-weight Quality Score upward.
• Split broad-match keywords into phrase or exact variants where intent is clear. Broad match without 30/30 conversions and Smart Bidding wastes spend on poor query matches.

4. Fix Ad Relevance Below Average — ad group splits + RSA rewrites

Ad Relevance rates how closely your keyword matches the message in your ad. Below Average means the RSA in the ad group is too generic to speak to the keyword's specific intent.

• Split the ad group by intent: one tightly-scoped concept per ad group (one product, one buyer stage, one match-type tier). Mixed-intent ad groups dilute Ad Relevance because no single RSA can speak to all queries it serves.
• Rewrite the RSA: add 12-15 unique headlines and 4 descriptions referencing the ad group theme. Each headline should be a different angle (benefit, proof point, CTA, feature, social proof).
• Avoid pinning unless brand or legal mandate. Pinned positions block Smart Bidding from picking the best combination for each auction. An acceptable pin is brand name in headline 1.
• For deeper RSA work see the RSA Ad Strength fix article.

5. Fix Landing Page Experience Below Average — page parity + speed

Landing Page Experience rates how relevant and useful your landing page is to people who click your ad [5]. Below Average means Google's automated systems and human raters scored the page poorly on relevance, transparency, or usability.

• Audit headline parity: the landing page H1 must mirror the ad headline language. If the ad promises "Free 14-Day Trial" the page H1 cannot say "Schedule a Demo".
• Measure mobile LCP (Largest Contentful Paint). Target under 2.5 seconds. Use PageSpeed Insights and Search Console Core Web Vitals to identify regressions.
• Verify HTTPS, mobile usability, visible trust signals (testimonials, security badges, privacy link), and clear navigation back to the main site.
• For the full landing page playbook see the Landing Page Experience fix article.

6. Re-export and re-measure after 21 days

Component ratings refresh slowly because Google needs auction data to recalculate them. Re-export the Quality Score diagnostic columns 21 days after the change is live and confirm:

• Below Average component count dropped to <30% of enabled keywords (rule status = passed).
• Median Quality Score lifted by 1-2 points on the previously failing cluster.
• Average CPC on those keywords fell at equal-or-higher CTR.
• Conversion volume held flat or grew — quality lift is only real if it is not bought via traffic loss.

Common mistakes

• Fixing the wrong component. Rewriting RSAs when the failing component is Landing Page Experience wastes the 21-day cycle. Always read the diagnostic columns first.
• Treating Quality Score as a single number. The 1-10 aggregate hides which component is failing. The component columns are where the actionable signal lives.
• Pinning every RSA headline. Pinning blocks Smart Bidding's combination optimization and frequently degrades Ad Relevance ratings. Reserve pinning for legal or brand mandates.
• Expecting same-week movement. Component ratings take 14-21 days of stable spend to refresh because they are auction-weighted, not snapshot-weighted.
• Pausing low-QS keywords before fixing. Pausing without component diagnosis loses the data that would tell you which lever to pull on the rest of the account. Diagnose first, then pause structural losses.

FAQ

Q: My Quality Score is 7 but Landing Page Experience reads Below Average. Should I still fix it?
Yes. The aggregate score hides component-level risk. A Below Average landing page is a future QS regression — Google's algorithm will downweight the keyword as auction data accumulates. Fix it before the aggregate drops.

Q: I have 5,000 keywords. Do I need to diagnose every one?
No. Focus on the keywords driving 80% of cost (the Pareto cluster) and the keywords with QS ≤ 4. Component fixes on the high-cost cluster move account-level metrics fastest.

Q: How does this rule relate to the keyword-level Quality Score rule?
The keyword-level Quality Score rule measures the share of keywords with QS ≤ 4 — the aggregate distribution. This component rule measures the share with at least one Below Average component — the diagnostic underneath. Fix this rule first because it tells you which lever to pull; the keyword-level rule then validates the fix moved the aggregate.

Q: Can I check QS components via the Google Ads API?
Yes. The keyword_view resource exposes quality_info.creative_quality_score, quality_info.post_click_quality_score, and quality_info.search_predicted_ctr as numeric ranks. The Whitead audit reads these directly. UI users see the same data via the four-column diagnostic above.

Q: What if Quality Score is null?
Google reports Quality Score as null until enough auction data accrues — typically 100+ impressions over 30 days. Filter out null-QS keywords before computing the failure share; they are not yet diagnostic.

Sources

1. Google Ads Help — About Quality Score for Search campaigns. https://support.google.com/google-ads/answer/6167118 (accessed 2026-05-26).
2. Google Ads Help — About Ad Rank. https://support.google.com/google-ads/answer/1722122 (accessed 2026-05-26).
3. Google Ads Help — Check your Quality Score. https://support.google.com/google-ads/answer/2454010 (accessed 2026-05-26).
4. Google Ads Help — Expected clickthrough rate. https://support.google.com/google-ads/answer/1659696 (accessed 2026-05-26).
5. Google Ads Help — About landing page experience. https://support.google.com/google-ads/answer/1659694 (accessed 2026-05-26).