Quality Score is a 1–10 diagnostic that Google Ads shows for each keyword. It is a summary of three signals (expected click-through rate, ad relevance and landing page experience), not the live number Google uses in each auction. It matters because the three signals behind it do influence what you pay per click and how often your ad shows. It matters less than most guides suggest because you cannot optimise the score itself; you can only improve what feeds it.
We look at Quality Score in the first hour of every account audit, but we never set it as a goal. Below is how we read it, what it will and will not tell you, and the three fixes that keep paying off on small accounts.
What is Quality Score, exactly?
Quality Score is a keyword-level rating from 1 to 10 that Google reports in the keyword table (you may need to add the column). Google has been clear in its own documentation that the number is a diagnostic tool meant to give you a rough sense of how your ads compare with other advertisers on that keyword. The real-time auction uses a more granular, query-by-query assessment that you cannot see.
The score is built from three components, each rated “below average”, “average” or “above average”:
- Expected click-through rate: how likely your ad is to be clicked when shown for that keyword, compared with advertisers competing on similar searches.
- Ad relevance: how closely your ad copy matches what the searcher is looking for.
- Landing page experience: how useful, relevant and quick the page behind the ad is for that search.
Those three signals feed into Ad Rank, which together with your bid decides whether your ad shows and in what position. Better signals generally mean you can win the same position at a lower bid. That is the commercial reason to care.
Does Quality Score still matter in 2026?
Yes, but with a narrower job than it had ten years ago. Two things changed. First, most accounts now run on automated bidding and broad or phrase match keywords paired with responsive search ads, so keyword-level tuning has less leverage than it used to. Second, Google increasingly evaluates intent query by query, so a single keyword can behave very differently across the searches it triggers.
What Quality Score is still good at is pointing to where a problem lives. If a keyword that spends well has “below average” for landing page experience, you have a page problem. If it shows “below average” expected CTR, your ad is not earning clicks against competitors. Treat it as a smoke alarm, not as a thermostat.
What Quality Score is not
Three misconceptions cost small advertisers money.
It is not your account’s performance. A keyword with a Quality Score of 4 that produces leads at €25 each is a better keyword than one with a score of 9 and no conversions. Conversions and cost per lead are the outcome; Quality Score is a hint about one input.
It is not the same as an Optimization Score. Optimization Score is Google’s percentage of recommendations you have applied. It is a different metric, and several of its suggestions raise spend rather than results. We cover how to treat recommendations in our Google Ads audit process.
It is not something you can improve by deleting low-score keywords. Pausing every keyword under 5 will raise your average score and will often cut volume and leads along with it. Only pause a low-scoring keyword if it also has poor cost per lead after enough data.
How do you find the real problem in 15 minutes?
Open the keyword view, add the columns for Quality Score, Exp. CTR, Ad relevance and Landing page exp., and filter to keywords with at least 100 clicks in the last 90 days. Smaller keywords give noisy ratings. Then sort by spend and look at the top ten.
For each keyword, find the weakest of the three components and write it down. In most small accounts we see one pattern dominate. A service business with one catch-all ad group tends to be weak on ad relevance. An e-commerce store sending all traffic to the homepage tends to be weak on landing page experience. An account that copies competitors’ generic headlines tends to be weak on expected CTR. Fix the dominant pattern first, because it usually affects most of your spend.
Fix 1: Tighten the ad groups so the ad matches the search
Ad relevance improves when the ad repeats the language of the search. If you sell boiler servicing, “boiler service” and “boiler repair” deserve separate ad groups with separate headlines, not one group that says “heating solutions” to everybody.
A practical rule we use: if two keywords would deserve different headlines or different landing pages, they belong in different ad groups. You do not need single-keyword ad groups. Three to ten closely related keywords per group is a workable size. Then make sure at least two or three headlines in the responsive search ad contain the core phrase. Pinning one headline to position 1 for the exact service name is a legitimate way to guarantee it appears, at the cost of some automatic variation.
Fix 2: Make the landing page earn its rating
Landing page experience is the component that most often stays stuck at “below average” because it needs work outside Google Ads. Three things matter in practice.
- Relevance: the page should clearly deal with the thing the ad promised. Sending “emergency plumber” clicks to a general services page is a mismatch. We describe the mechanics in our post on landing page and ad message match.
- Speed: a page that takes several seconds to become usable on a mobile connection loses people before they read a word. Compress images and remove unused scripts before you touch anything else.
- Clarity and trust: a visible price or “from” price where possible, a phone number or form above the fold, and real proof such as reviews or accreditations.
After you change a page, give the rating a few weeks. The landing page component does not update instantly, and it needs enough traffic to be re-evaluated.
Fix 3: Write ads that earn clicks against the competition
Expected click-through rate is relative. You are being compared with others bidding on the same searches, so generic copy such as “Quality service, call today” loses to ads with a specific number, a clear difference or an offer. Our reasoning is simple: the searcher scans four or five ads in two seconds, and the ad that answers their exact question wins the click.
What works on small accounts, in our experience: a headline with the searched phrase, a headline with a concrete proof point (years in business, a rating, a guarantee), and a headline with a call to action that matches the page. Add all available assets, including sitelinks, callouts and a structured snippet, because they enlarge the ad and give more reasons to click. Then let the ads run long enough to judge. A rough guide is at least a few hundred impressions per ad variant before you decide anything, and several weeks for low-volume accounts.
A quick comparison: where each fix pays most
| Weak component | Typical cause | First fix | Time to see effect |
|---|---|---|---|
| Ad relevance | Mixed keywords in one ad group | Split ad groups, repeat keyword in headlines | Days to two weeks |
| Expected CTR | Generic copy, few assets | Specific headlines, add all assets | Two to four weeks |
| Landing page experience | Mismatch, slow page, thin content | Dedicated page per service, speed work | Several weeks |
What a realistic improvement looks like
We will not give you a promised percentage, because the effect depends on the market and on how bad things were. As a way to reason about it: if better signals let you hold the same position at a bid that is, say, 15 per cent lower, an account spending €2,000 a month would free up about €300 a month to buy more clicks. That figure is an illustration of the arithmetic, not a benchmark. The point is that the gain shows up as either lower cost per click or more impression share at the same budget, and you verify it in your cost per lead, not in the score.
When to ignore Quality Score entirely
Ignore it for keywords with fewer than about 100 clicks, for brand terms (these usually score high anyway), and in the first two weeks of a new campaign when data is thin. Also ignore it when the keyword converts profitably. A profitable keyword with a mediocre score is a candidate for gradual improvement, not for panic.
If you are running Performance Max or other automated campaign types, there is no keyword-level score to read at all. The same three questions still apply, though: does the creative match the intent, do people click, and does the destination deliver? Our piece on Target ROAS bidding on a small budget shows how those automated systems depend on the same fundamentals.
FAQ
What is a good Quality Score?
Seven or above is usually healthy and five to six is workable. Scores below five on a high-spend keyword deserve investigation, but only if that keyword is also performing poorly on cost per lead.
Can I raise my Quality Score by bidding more?
No. Bids affect position and cost, not the three quality signals. Quality Score reflects ad relevance, expected click-through rate and landing page experience.
How long does it take for Quality Score to update?
Changes to ads and keywords can show up within days if there is enough traffic. Landing page changes often take several weeks to be reflected.
Does Quality Score matter for Performance Max?
There is no keyword-level score in Performance Max, but the same principles of relevant creative and a good destination page still drive results.