Google Ads Quality Score — How It Works, How to Track It, How to Fix Drops

What QS actually is

Quality Score is Google’s 1-10 estimate of how relevant your ad + landing page is to a user’s search query. It influences ad rank in real-time auctions: a higher QS lets you outrank a higher-bidding competitor, which means you pay less for the same impression.

It’s computed per-keyword (not per-campaign) and updates daily based on accumulating performance data. New keywords usually start at 5-6 and find their level within 3-4 weeks of meaningful impression volume.

The three sub-scores, decoded

QS is the headline number. The three sub-scores are the diagnostic:

1. Expected CTR. Google’s estimate of how often users would click your ad if it were shown for this query, normalized for ad position. Below average means you have a click-through-rate problem — usually weak ad copy, irrelevant keywords, or position-hostile ads (like price-anchored ads in #4 position).
2. Ad relevance. How tightly your ad text matches the keyword. Below average almost always means the keyword is in an ad group with off-topic ads. Single-keyword or single-theme ad groups solve this immediately.
3. Landing page experience. Combines load speed, mobile-friendliness, content-keyword match, and policy compliance. Below average usually means slow landing pages or generic homepage destinations for specific keywords.

Each sub-score is rated Above average, Average, or Below average. Aim for all three at Average or better; that produces QS 7-8 minimum.

How to track QS over time

Default Google Ads UI doesn’t show QS history. To track:

• Manual: export keywords + QS daily/weekly to a spreadsheet. Tedious; works.
• Scripts: Apps Script can pull QS via the AdWords API and log to a Sheet.
• Tooling: Optmyzr, Adalysis, and MarqOps all track QS history natively. MarqOps shows median QS sparklines per keyword and fires drop alerts when median QS falls 2+ points in 7 days.

Why daily tracking matters: QS regressions usually start as quiet drift (8 → 7 → 6 over three weeks), not sudden cliffs. By the time the drop is visible in monthly reports, you’ve been bleeding CPC efficiency for weeks.

Why QS drops happen

The five common patterns:

1. Match-type expansion. Phrase or broad keywords pull in queries the ad doesn’t match → expected CTR + ad relevance both drop.
2. Landing-page regression. A site speed issue (image bloat, third-party script) pushes mobile load time past 3-4s → landing page experience drops.
3. Ad-copy disapproval. One RSA variant disapproves silently and the rotation collapses to lower-performing variants → expected CTR drops.
4. Competitive surge. A new competitor enters the auction with sharper copy → your relative CTR drops, even though absolute CTR holds.
5. Ad-group dilution. Adding new keywords to an existing ad group makes the existing copy less relevant to all of them → ad relevance drops across the board.

The fix-it playbook

If you have a QS problem, work in this order — earlier steps are highest-leverage:

1. Pause QS ≤ 3 keywords. Anything below 4 with meaningful clicks is a money pit. Add as exact-match negatives if the query is genuinely off-target.
2. Tighten ad groups. Single-theme groups with 3-5 closely-related keywords each. Each ad group gets ads that explicitly mention the theme.
3. Improve landing-page speed. Mobile LCP under 2.5s, CLS under 0.1, INP under 200ms. The free audit measures all three.
4. Match landing copy to keyword. Generic homepage = below-average landing page experience for specific commercial-intent keywords.
5. Run 3 RSA variants per ad group. One disapproval doesn’t cripple expected CTR if you have variety.
6. Use ad customizers for headline 1. Insert the keyword into the headline; expected CTR climbs measurably.
7. Track + alert. So the next regression doesn’t happen unnoticed. MarqOps’s QS trend tracker is built for this.

Tactical sequence aside: QS 8-10 keywords usually pay 30-50% less per click than QS 4-6 keywords for the same ad rank. Closing the gap on your top-spend keywords is a 3-month project that often pays for the entire PPC tooling budget.