Ad Strength scores creative diversity, not quality — over-pinning headlines is the #1 cause of a Poor rating, because each pin slashes testable combinations.
What Poor actually means
Ad Strength rates a Responsive Search Ad (RSA) on a 4-step scale — Poor, Average, Good, Excellent — based on how many distinct, relevant headline and description combinations Google can assemble from your assets (Google Ads Help, 2026-01-22). The score updates the moment you save the ad and rechecks as Google learns. A Poor rating tells you that the system either cannot generate enough variations, finds the variations it can generate too repetitive, or sees an ad whose headlines are nearly synonymous. Google recommends supplying the maximum 15 headlines and 4 descriptions, but the rating is calculated on diversity and relevance — not the raw asset count.
Why pinning kills the score
Pinning forces a headline or description into a specific slot (1, 2, or 3). Each pin tells the system: "for this slot, only consider these assets." Optmyzr's combinatorial analysis shows pinning a single headline to position 1 collapses possible serving combinations from roughly 43,680 down to about 10,920 — a 75% reduction (Optmyzr, 2025-08-12). Pin three headlines and you have effectively built a fixed expanded text ad with extras. Google's 2024 unpinning guidance was direct:
"We recommend not pinning headlines or descriptions, so we can find the best-performing combinations for your business." (Google Ads Help: About RSAs, 2024-06-04)
If legal or brand requires pin-slot 1, pin multiple assets to the same slot rather than one — Google then rotates among them and Ad Strength typically recovers from Poor to at least Average.
Headline repetition and similarity
Even unpinned RSAs score Poor when the system flags headlines as too similar. Google's 2024 update added stricter near-duplicate detection: variants like Buy Running Shoes, Order Running Shoes, Running Shoes Online, and Shop Running Shoes register as one idea repeated four times. The Improve Ad Strength panel typically surfaces this as Add more unique headlines even when the asset count is at 15. Aim for headlines that cover distinct angles: brand, USP, offer, social proof, urgency, feature, audience, location. The 8-10 unique angle heuristic from Search Engine Land's 2025 RSA teardown is a useful floor — fewer than that and the diversity check trips (Search Engine Land, 2025-10-08).
In Whitead audits, when we find clusters of Poor Ad Strength, the root cause sits in two patterns roughly 80% of the time — and almost never in "not enough assets". Pattern one: legacy ETA migration where the original 3 headlines were pinned to preserve old wording, then 12 generic fillers added to satisfy the asset-count nudge. Pattern two: one brand-voice template ("Buy X Today", "Order X Now", "Get X Fast") cloned across 15 slots so token overlap is enormous. Both score Poor for the same reason: the system has nothing new to test. The fix is not "write more headlines" — it is "write headlines that say different things". This usually moves the account from Poor to Good within one edit cycle.
How to recover from Poor
Work the Improve Ad Strength side panel top-down: it lists the missing levers in priority order. First, unpin everything you do not legally need. Second, replace near-duplicates with angle-diverse alternatives — use the search-terms report to source language users actually type. Third, fill genuine gaps: if Ad Strength flags Add more headlines including popular keywords, paste the top 3 query themes from the last 30 days. Save the ad, wait 10-15 minutes for the score to refresh, and recheck. Do not chase Excellent at the expense of brand voice — Google's own data shows the lift from Poor → Good is far larger than Good → Excellent, and conversion-rate impact tracks quality-score signals more closely than the Ad Strength badge itself. If Poor persists after a clean rewrite, sanity-check that the ad-group keywords match the headline copy at all — a mismatch will also drag search-ctr-low.
Does Poor Ad Strength actually reduce impressions?
Yes, in practice. Google states Ad Strength is a diagnostic, but its 2022 internal study reported advertisers who improved Ad Strength from Poor to Excellent saw an average +12% conversions (Google Ads Help, 2026-01-22). The auction itself uses Ad Rank, but Poor ads have less creative inventory to serve, so they win fewer impressions over time.
Should I pin headlines to control brand or legal copy?
Pin to a slot and stack 2-3 approved alternates in the same slot. That preserves compliance while letting Google rotate within the allowed pool. A single asset pinned alone is the worst case — it locks a slot and signals low diversity. The unpinning guidance from 2024 specifically calls out the multiple-assets-per-slot workaround as the recommended pattern.
Is Excellent worth chasing on every ad?
No. Google's own benchmarks show diminishing returns past Good. Optmyzr's 2025 analysis of >50,000 RSAs found Good and Excellent ads converted within 2-3% of each other, while Poor underperformed Good by double digits (Optmyzr, 2025-08-12). Spend the optimisation budget on lifting Poor ads, not polishing Good ones to Excellent.
Related concepts
See Quality Score for how relevance signals affect the auction independently of the Ad Strength badge, and Search CTR Low for diagnosing whether weak ads or weak keyword-to-copy fit is the bigger lever.
Sources
- About Ad Strength — Google Ads Help, 2026-01
- About responsive search ads — Google Ads Help, 2024-06
- The math behind RSA pinning and Ad Strength — Optmyzr, 2025-08
- Responsive search ads best practices — Search Engine Land, 2025-10