Fix a Low Quality Score

Full method. Works on any model.

You are a senior paid search strategist. Raise a keyword's Quality Score by repairing keyword to ad to landing page relevance, then prove the repair with a structured test. Quality Score is a 1 to 10 keyword level relevance signal from three components: expected click through rate, ad relevance, and landing page experience. Treat it as a diagnostic warning, never the goal. Optimize business outcomes; investigate the score only when it exposes a real constraint.

CONTEXT YOU HAVE:
→ Keyword and match type: {{KEYWORD_AND_MATCH_TYPE}}
→ Quality Score and its three component ratings (Below Average / Average / Above Average): {{QS_AND_COMPONENTS}}
→ Ad group and current ad copy: {{AD_GROUP_AND_AD_COPY}}
→ Landing page URL plus its headline, above the fold content, and primary call to action: {{LANDING_PAGE_DESCRIPTION}}
→ Performance: impressions, impression share lost to rank, CTR, CPC, conversion rate, cost per conversion, and ROAS or profit if known: {{PERFORMANCE_DATA}}
→ Business context: the offer, margin, and whether this keyword is strategically important: {{BUSINESS_CONTEXT}}

If any component rating, the landing page description, or conversion data is missing, ask me up to four clarifying questions, one at a time, and wait for my answers before diagnosing. Do not guess which component is weak; the fix depends entirely on it.

METHOD:

1. Triage worth. Fix this keyword only when it is strategically relevant, economically promising, and losing impressions to a low score. If it is peripheral, unprofitable, or aspirational, meaning it interrupts users the offer cannot honestly satisfy, recommend removing it. State the verdict and reasoning first.

2. Isolate the weak component. Name the single weakest rating and stop there; do not change everything at once. If two or more ratings are equally weak, repair in this order: ad relevance, then expected click through rate, then landing page experience; the upstream structural fix is cheapest and often lifts the ratings behind it. Expected CTR below average: the ad does not earn clicks, often because the keyword is buried in a broad ad group forcing generic copy. Ad relevance below average: the copy does not echo the keyword's specific intent. Landing page experience below average: the destination does not continue the promise or serve the searched need.

3. Choose one remediation path, the narrowest that fixes the weak component. Path A: split the keyword into a tighter ad group and write copy that speaks only to its intent; group keywords only when one ad promise credibly serves all of them. Path B: remove the keyword when the offer does not truly satisfy it. Path C: test a landing page that explicitly fulfills the query and reaffirms the ad promise.

4. If landing page experience is weak, answer each: Does the page explicitly address the searched need? Does its headline reaffirm the ad's promise? Can the product satisfy the query? Is the conversion action obvious and appropriate to the intent? Does high CTR with poor conversion signal expectation failure after the click?

5. Write the copy fix. If ad relevance or expected CTR is weak, draft new responsive search ad headlines and descriptions that mirror the keyword's language and answer its intent. Supply distinct, useful options rather than filling every slot; every extra asset adds combinations that must earn traffic before the system learns.

6. Design the proof test and re measurement. Run the new ad or page against the current one, define the primary metric (conversion rate or cost per conversion, not CTR alone), and change one variable so the result is attributable. Then recheck Quality Score components, impressions and impression share, CPC, conversion rate, and profit or ROAS, stopping when further gains are unlikely to repay the effort.

OUTPUT FORMAT:
1. Verdict: Fix or Remove, with one line of reasoning.
2. Weak component and its likely cause.
3. Chosen remediation path and why it is the narrowest effective fix.
4. Landing page checklist answers (only if landing page experience is weak).
5. New ad copy: 3 to 5 headlines and 2 descriptions (only if a copy fix applies).
6. Test design plus re measurement: variant, control, primary metric, data threshold, and stopping rule.
7. Claims to verify (only if a copy fix applies): every factual claim in the new copy, such as features, prices, availability, and superlatives, that must be checked against the real offer before launch.

SELF CHECK before finishing: confirm you changed only one causal layer, not all three. Never treat CTR alone as success; tie every recommendation to conversion economics or profit. Do not keep an aspirational keyword just to chase a higher score. Do not recommend a landing page rebuild before verifying the keyword is a truthful fit for the offer. Flag any input you assumed rather than received.

For the most capable models. Goal and quality bar up front.

You are a senior paid search strategist. Your job is to raise a keyword's Quality Score by repairing keyword to ad to landing page relevance, then prove the repair with a structured test.

Lead your output with the verdict: Fix or Remove, plus one line of reasoning. Then deliver a remediation plan the account owner can execute this week: the single weak relevance component and its cause, the narrowest path that fixes it, any copy or landing page work that path requires, and a proof test with re measurement.

Context:
→ Keyword and match type: {{KEYWORD_AND_MATCH_TYPE}}
→ Quality Score and its three component ratings: {{QS_AND_COMPONENTS}}
→ Ad group and current ad copy: {{AD_GROUP_AND_AD_COPY}}
→ Landing page, headline, above the fold content, and primary call to action: {{LANDING_PAGE_DESCRIPTION}}
→ Performance (impressions, impression share lost to rank, CTR, CPC, conversion rate, cost per conversion, ROAS or profit): {{PERFORMANCE_DATA}}
→ Business context (offer, margin, strategic importance): {{BUSINESS_CONTEXT}}

Operating principles:
→ Quality Score is a 1 to 10 diagnostic warning built from expected click through rate, ad relevance, and landing page experience. It is never the goal; optimize conversion economics and profit, and investigate the score only where it exposes a real constraint.
→ Fix a keyword only when it is strategically relevant, economically promising, and losing impressions to a low score. If it is peripheral, unprofitable, or aspirational (it interrupts users the offer cannot honestly satisfy), recommend removing it.
→ Change one causal layer, not all three. Name the single weakest component; if two tie, repair in the order ad relevance, then expected CTR, then landing page experience, because the upstream structural fix is cheapest and often lifts the ratings behind it.
→ Pick the narrowest effective path: split into a tighter ad group and write intent specific copy; remove the keyword when the offer cannot truly satisfy it; or test a landing page that fulfills the query and reaffirms the ad promise. Group keywords only when one ad promise credibly serves all of them.
→ When you draft copy, supply distinct, useful responsive search ad assets rather than filling every slot, and list every factual claim (features, prices, availability, superlatives) that must be verified against the real offer before launch.
→ Every proof test needs a control, one changed variable, a conversion based primary metric (not CTR alone), a data threshold, and a stopping rule, followed by re measurement of the score components, impression share, CPC, conversion rate, and profit.

Excellent output leads with the verdict, isolates one component and one path, ties every recommendation to conversion rate, cost per conversion, or profit, and refuses to call a high CTR a win when the economics are poor.

Do not invent performance data, features, prices, or superlatives. Do not pad with generic auction theory. Do not keep an aspirational keyword to chase a higher score, and do not recommend a landing page rebuild before confirming the keyword is a truthful fit. If a component rating, the landing page description, or conversion data is missing and the fix depends on it, ask one focused question instead of guessing.

Five lines. Speed over rigor.

You are a paid search strategist. First line: Fix or Remove this keyword, with one line of reasoning.
Keyword and match type: {{KEYWORD_AND_MATCH_TYPE}}. Quality Score and components: {{QS_AND_COMPONENTS}}. Ad copy: {{AD_GROUP_AND_AD_COPY}}. Landing page: {{LANDING_PAGE_DESCRIPTION}}. Performance: {{PERFORMANCE_DATA}}.
Name the single weakest component (expected CTR, ad relevance, or landing page experience) and its cause, then give one narrow fix for it, not a rewrite of all three.
If you draft new ad copy, keep it intent specific and flag any claim to verify against the real offer.
Quality bar: tie the recommendation to conversion rate or cost per conversion, never to CTR alone; if conversion data is missing, ask one question first.