Quality Score and ad strength at a glance
Quality Score is reported for each keyword on a scale from one to ten, with three components beneath it: expected click-through rate, ad relevance and landing page experience, each marked Above average, Average or Below average against other advertisers on the same searches. It is an estimate aggregated over time, so it moves slowly, and a keyword with too few recent impressions shows no score at all. The Quality Score guide explains each component, how the score relates to Ad RankThe value Google Ads calculates at each auction to decide if and where an ad shows. and the price of a click, and which component to fix first.
Ad strength belongs to a single responsive search ad. It runs from Incomplete through Poor, Average and Good to Excellent, and it appears while the ad is being built. Google describes it as a measure of the relevance, quality and diversity of the ad's assets, and attaches suggestions: add more headlines, include popular keywords, make headlines or descriptions more distinct, unpin assets. Because it is calculated from the assets themselves, a brand-new ad has a rating before it has served once. The responsive search ads guide covers how the ad is assembled and how to write a pool of assets that combine well.
Both ratings sit outside the auction. Google's help page on Quality Score states that the score is not an input to Ad Rank, and its help page on responsive search ads states the same of ad strength. What the auction uses is its own assessment of the ad's quality for that search, at that moment, from the same ingredients the keyword score summarises. So the two ratings are signals about the auction, not levers on it, and they answer different questions: Quality Score asks how a keyword's traffic has been received; ad strength asks whether the ad gives the system enough good material to work with.
Compare table
Quality Score vs Ad strength
| Row label | Quality Score | Ad strength |
|---|---|---|
| What is rated | A keyword: how its ads and landing page have fared for the searches it matched | One responsive search ad: the completeness, relevance and variety of its assets |
| Scale | One to ten, with three components rated Above average, Average or Below average | Incomplete, Poor, Average, Good or Excellent, with suggestions |
| When it is calculated | From recent auctions, aggregated over time; no score until the keyword has enough traffic | From the assets as the ad is written, before and regardless of serving |
| What feeds it | Expected click-through, ad relevance, landing page experience | Number and distinctness of headlines and descriptions, use of keywords, pinning |
| Place in the auction | Not an input; a summary of the quality factors the auction assesses each time | Not a factor; the auction uses predicted quality, not the rating |
| Link to cost | Indirect but real: weak components mean paying more for the same position | None documented; a better pool can lift results, which is what changes cost |
| Where it is shown | Keyword columns, with historical columns for dating a change | The ad editor and the ads table |
| What it suggests | Which of three components is weak, and so where the fix lies | Which writing gap to close: more assets, more variety, keywords, fewer pins |
| How fast it moves | Slowly, as new auctions replace old ones | Immediately, as assets are added or changed |
| Typical misuse | Reporting an account average as a headline figure | Padding the pool with empty headlines to reach Excellent |
The rows that matter most are the first and the fifth. Quality Score describes traffic that has already happened, so it can tell you about the landing page and the searcher's response, which ad strength cannot see. Ad strength describes material that exists before any traffic, so it is the only one of the two that can guide the writing of a new ad. Neither is consulted by the auction.
When to act on Quality Score
Act on Quality Score when a keyword that matters is paying more per click than it should, or losing impressions because its rank is too low. The components say why. Below-average expected click-through usually means the ad does not answer what people searching that term want; below-average ad relevance means the ad group's messages do not reflect its keywords; below-average landing page experience means the page does not deliver on the click. Each has a different fix, and the Quality Score guide's diagnosis table matches component to cause. Where the visible symptom is a rising price rather than a low score, the rising CPC diagnosis separates quality from competition, bidding and traffic mix.
Landing page experience is the component ad strength can never reveal, because ad strength never looks past the ad. A keyword whose ads rate Excellent can still carry a below-average landing page rating, and the fix is on the page: making it answer the query the ad promised, which is what message match means in practice. The CPC calculator shows what a lower average price would do to the clicks a budget buys.
Leave the score alone where chasing it would cost more than it saves. Competitor keywords rate poorly almost by design, discovery keywords on broad match carry low scores while they find new searches, and brand keywords usually rate well without effort. The comparison of broad and exact match explains why looser matching brings searches the ad cannot all fit.
When to act on ad strength
Act on ad strength while writing and when an ad has no results yet. For a new ad, the rating and its suggestions are the best guide available: too few headlines, near-duplicates, keywords missing from the headlines and heavy pinning all leave the system little to combine, and fixing them is cheap. Building the pool to Good or better is a sensible floor, as the responsive search ads guide recommends.
Stop acting on it when the ad has a record. An established ad is judged on its click-through, conversion rate and cost per result, and the ad strength diagnosis starts by putting the rating beside those figures. If the results are weak and the rating is Poor, the suggestions usually name the cause. If the results are the best in the group, the rating is measuring something other than what the account pays for. Reaching Excellent by adding headlines that say nothing new, or by unpinning a line that must appear, makes the ad worse to satisfy a rating the auction does not read.
There is one point where the two ratings meet. The suggestion to include popular keywords in headlines answers the same gap that the ad relevance component of Quality Score detects. Writing several headlines that use or paraphrase the ad group's keywords is the rare change that can move both.
A locksmith reads both ratings before rewriting its ads
Trindoux Locksmiths, an invented locksmith, runs two search ad groups. In "emergency lockout" its ad rates Excellent ad strength, yet its six main keywords score 4 out of 10, with landing page experience Below average and the other two components Average. Clicks cost £6.80 on average and 31% of impressions are lost to rank. In "lock change" the only ad rates Poor, with a suggestion to add headlines, yet it has the best click-through in the account at 9.2% and a cost per booked call of £18, against £34 for lockouts. The owner had planned to rewrite the lock change ad. Instead, the lockout landing page is rebuilt to lead with the response promise and a phone number above the fold, matching what the ad says. Over the following weeks landing page experience moves to Average, the keywords reach 6, the average click price falls to £5.40 and lost impression share to rank drops to 19%. In lock change, two new headlines that use the keywords are added alongside the existing ones; ad strength moves to Average and the ad's results hold.
Decision table
| Situation | Read | Why |
|---|---|---|
| Writing a new responsive search ad | Ad strength and its suggestions | No results exist yet; the rating checks the pool is complete |
| A keyword's clicks cost more than similar keywords | Quality Score components | The weak component names the fix |
| An established ad rates Poor but has the best results | The results, then the suggestions | The rating estimates; the results are what you pay for |
| Excellent ad strength, low Quality Score | Landing page experience and expected click-through | Ad strength cannot see the page or the searcher's response |
| Rising cost per click across a campaign | Quality Score history, then the CPC diagnosis | Separates a quality slip from competition and bidding |
| Keyword shows no score | Its traffic, not its quality | Too few recent impressions to rate |
| Reporting quality to a client or manager | Components on keywords that matter, with results | Account averages of either rating mislead |
When the two ratings disagree
Disagreement is normal, because the ratings look at different things. Excellent ad strength beside a low Quality Score usually means the ad pool is complete but something outside it is weak: the landing page, or the match between the keywords and what searchers want. Poor ad strength beside a healthy Quality Score usually means a short or repetitive pool that searchers nevertheless respond to, often because one or two headlines carry the message. Neither case calls for chasing the weaker rating; each calls for reading the component or suggestion that names a specific gap and checking it against results.
When click-through is the result in question, the CTR calculator compares ads on the same basis, and the comparison of CTR and conversion rate explains why an ad that wins the click has still to deliver the conversion. The search campaigns hub places both ratings among the other parts of a search campaign, and the comparison library lists the other pairs that are easy to confuse.
Common mistakes
- Treating either rating as an auction input. Google documents that neither is; the auction assesses quality for each search.
- Padding ads to reach Excellent. Headlines that repeat an idea or say nothing lift the rating and weaken the ad.
- Reporting an account average Quality Score. The score exists per keyword; averages hide which keyword and which component need work.
- Rewriting a winning ad because it rates Poor. Test a fuller version beside it instead.
- Expecting ad strength to fix a landing page problem. Landing page experience is outside the ad.
- Chasing scores on competitor and discovery keywords. Low ratings there are expected; judge the cost per result.
- Reading a change in one rating as a change in results. Confirm in click-through, conversions and cost before acting on either.
