Quality Score

Two brands bid on the same keyword. One pays less per click and shows higher on the page. Neither gap is always about who bid more.

Google assigns every search keyword a score from 1 to 10 for how relevant your ad and landing page are to what the person typed. It is built from three signals: expected click-through rate, ad relevance, and landing page experience. Raise the score and, in most accounts, the cost-per-click drops.

There is a caveat most advertisers skim past. The visible 1-10 figure “is not an input in the ad auction,” Google’s own Search Quality Score documentation states — it calls the number a diagnostic. The rest of this entry is about why that line is true and why your CPC still moves with the score anyway.

Why Little Sleepies writes 86 headlines instead of five

Little Sleepies runs 89 search ads on Google. 86 of the headlines are unique, per a March 2026 teardown of the account. The brand is not running five headlines and testing variations against each other.

The reason sits inside the ad-relevance component. A headline that repeats the words in the query scores higher on relevance, and a higher score buys a lower price on that click.

A parent searching “bamboo toddler pajamas for eczema” gets a headline built from those exact words. “Bamboo” shows up in 47% of the search headlines.

Our Place runs the same play at larger scale. Of 425 Google text ads, 111 of the cleanly extractable headlines are different from each other, per its teardown. The account matches “air fryer without PFAS” to “Non-Toxic Airfryer” and “titanium non stick pan” to “The First No-Coating Nonstick.”

Both brands are buying a lower price on each query by making the ad read like the search. The headline volume is a Quality Score tactic wearing the costume of creative testing.

The number you see is not the number that bids

The 1-10 score in the Google Ads interface is a stored, historical figure. Auction-time quality is recalculated for every individual query. That real-time value, not the digit sitting in your account, is what feeds Ad Rank, per a 2026 Optmyzr analysis.

This is the reconciliation behind Google’s “not an input” language. The signals inside Quality Score do shape what you pay. The single visible digit is a lagging summary of them, which is why two keywords showing the same score can clear the auction at different prices.

For a D2C account, the stored score is still the cheapest diagnostic on the shelf. A keyword sitting at 4 while its neighbours sit at 8 is flagging an ad-relevance or landing-page gap on that specific term, before any spend data comes in.

The same mechanism, in reverse: competitor terms

Bid on a competitor’s brand name and the relevance signal collapses. Your ad and landing page are about your product, not theirs, so the ad-relevance and landing-page components score low. Competitor keywords typically carry Quality Scores of 2-5 and cost 2-5 times more per click than your own brand terms, per Free Agency’s conquesting guide.

The teardown corpus shows how rarely D2C brands accept that premium. Fresh Clean Threads runs exactly one named-competitor ad — “Fresh Clean Tees Vs True Classic” — still active on 13 March 2026 in its Google account. Out of 543 ads, it is the only one paying the low-relevance tax on purpose.

Where Quality Score stops existing

Quality Score is a Search-campaign metric. It does not exist for Shopping ads, and it does not exist for Performance Max. Google assembles those placements from a product feed and its own signals, with no keyword and no 1-10 relevance figure to read.

That gap matters because Shopping is where much of the D2C search budget sits. 60% of Little Sleepies’ 524 Google ads are Shopping, by the teardown’s count. For that majority of the account, the lever this entry describes is not present, and feed quality takes its place.

The score answers a narrow question about text search ads. The half of the account running on feeds answers to something else entirely.