How to Check Quality Score in Google Ads and Improve It
Table of contents
Twenty seconds. That is how long it takes to make Quality Score appear in Google Ads, and until you do, it appears nowhere. It lives in the Keywords view, behind the Columns menu, and Google leaves it switched off by default.
Finding the number is the easy half. Reading it properly, and knowing which of its three parts to fix, is where accounts go wrong. This guide walks through both, in the order you would do them in the account.
One piece of context first. Google’s help page says “Quality Score is not an input in the ad auction”; it is a diagnostic that summarises the ad quality the auction does assess, and the Ad Rank guide covers how that auction works.
How to check Quality Score, step by step
1. Add the four Quality Score columns
Open Campaigns, then the Keywords page, and click the columns icon at the top right of the table. Choose Modify columns, expand the Quality Score section, tick all four columns and hit Apply:
- Quality Score gives you the 1 to 10 figure for each keyword.
- Exp. CTR shows how likely Google thinks your ad is to be clicked.
- Ad Relevance shows how closely your ad matches the search.
- Landing Page Exp. shows how relevant and useful your destination is.
Add all four together, because the score on its own tells you nothing about what to fix. Then save the column set. Column layouts reset between views, and rebuilding this on every visit is the reason so many advertisers check it twice and never again.
Quality Score is reported per keyword only. There is no campaign or ad group figure, though you can sort or filter the keyword view to compare groups. For anything above a few hundred keywords, use the download button above the table and work in a spreadsheet, or pull the same columns in Google Ads Editor if you prefer working offline.
2. Add the historical columns too
The same section offers Quality Score (hist.), Exp. CTR (hist.), Ad Relevance (hist.) and Landing Page Exper. (hist.). The standard columns show only the current value, so you cannot tell whether a keyword scoring 5 has climbed from 3 or fallen from 8.
With the historical columns added, the figures follow your date range. Set the range to the month before a change, note the figure, then set it to the month after. Without that comparison you are guessing whether a rewrite helped, and guessing is how accounts end up rewritten three times for no gain.
3. Sort by cost, not by score
Sort the table by cost, descending, and ignore everything with no spend. The keywords burning your budget at a 4 are the entire story. The ones sitting at 3 with no spend are noise.
Resist averaging the scores across the account. Google warns that Quality Score “is not a key performance indicator and should not be optimized or aggregated with the rest of your data”. A plain average weights a keyword that spent AUD 4,000 the same as one that spent nothing. If you want one account-level figure, export Quality Score and Cost, multiply each score by its cost, total that column and divide by total cost. That cost-weighted figure usually sits well below the plain average.
4. Read the score as a summary
Ten means your keyword, your ad and your landing page are all strongly aligned with what people are typing. One means at least one of those three is badly out of step.
As a working rule, treat 5 or below as a real problem, 6 and 7 as normal, and 8 and above as fine. Stop working on anything at 8 or more; the effort is better spent elsewhere.
5. Find the component marked Below average
The three components report as Below average, Average or Above average rather than numbers. Those three words are the instruction. The 1 to 10 score only summarises them.
Each status compares you with other advertisers whose ads showed for the exact same search over the last 90 days. That is why a score can drop while your account stays still: a stronger competitor entering your auctions, or a seasonal shift in searches, changes the benchmark you are measured against.
6. Understand what a dash means
A dash in the column means Google has not collected enough data. In Google’s wording, it appears when “there aren’t enough searches that exactly match your keywords” to determine a score.
This is normal for new keywords and permanent for very low-volume ones, which for a suburb-level Australian search term may simply mean it never fills in. A dash is neither a bad score nor a warning, so judge those keywords on cost per conversion instead. If most of your account shows dashes, the cause is usually too many hyper-specific keywords, and the fix is a shorter, better-chosen keyword list rather than more of them.
You will also find no Quality Score at all outside Search. Shopping, Performance Max, Display and Video campaigns have no keywords to attach one to. In a Shopping account, the equivalent work happens in your product feed.
The three components and what to fix first
Google defines each component on its own help page: expected clickthrough rate, ad relevance (defined in Google’s Quality Score overview as “how closely your ad matches the intent behind a user’s search”) and landing page experience. The table turns those definitions into a starting point for each Below average rating.
| Component | What it measures | What “Below average” usually points to | First fix |
|---|---|---|---|
| Expected CTR | How likely your ad is to be clicked when shown for that keyword, adjusted to remove the effect of ad position and assets | Generic ad copy, or a keyword that has drifted away from what the ad offers | Rewrite the ad around a concrete offer: a price, a timeframe, a guarantee or a location |
| Ad relevance | How closely your ad matches the intent behind the search | An ad group carrying several intents, so no single ad can match them all | Split the ad group, then put the searcher’s words in the headlines |
| Landing page experience | How relevant and useful the page is to people who click the ad | Traffic sent to the homepage or a shared page, or a slow page on mobile | Send each ad group to a page built for that search and check its mobile load speed |
Fixing expected CTR
Rewrite the ad around the offer rather than the business. Google’s advice on using Quality Score is to “edit your ad text to make your offer more compelling to your target audience” and to “ensure the details in your ad match the intent of your keywords”.
Concrete beats clever. A price, a timeframe, a guarantee or a location will usually outperform an adjective, which is why ad copy is closer to commercial writing than to design work.
Expected CTR tends to respond first, because it rebuilds as new impressions accumulate. That makes it the sensible place to start if you want to see whether your changes are landing at all.
Fixing ad relevance
This component is more sensitive to structure than to writing. Google’s instruction is to “match the language of your ad text more directly to user search terms”, and where one ad group holds keywords for different searches, to split it into several groups that better match them.
A quick test: if you cannot write one headline that suits every keyword in the ad group, the ad group is the problem. Split it before you rewrite anything.
Fixing landing page experience
This is the slowest of the three to improve and the one that drags longest when broken. It is also the only component that is not really an advertising problem, so a Below average rating here often calls for work on the page itself rather than another round of ad testing.
Send each ad group to a page built for that search, not to your homepage. Then check the page loads quickly on a phone over mobile data, since Google’s guidance names mobile friendliness and loading speed, and a large share of Australian search traffic arrives that way. A page that already earns its place in organic search results will often score well here too.
Two failures account for a large share of Below average ratings: several ad groups pointed at the same page, so none of them match closely, and a page that loads a full-width hero image before anything useful appears. Both are fixable without touching the ads.
What order to work in
Start with the expensive keywords reading Below average on two or more components, because that is where your money is going. Fix ad relevance and expected CTR first since they respond soonest, and start the landing page work straight away because it takes longest to be reassessed.
Keep Quality Score and Ad Strength apart while you do this. Quality Score rates a keyword on click likelihood, relevance and landing page; Ad Strength rates an ad on how many asset combinations it offers. A keyword can hold a 9 behind an ad rated Poor, and the reverse is common. Whether Ad Strength is worth chasing is a separate question.
If your account is large enough that this sorting exercise takes a full day, that is usually the point where handing the account over costs less than doing it yourself.
As for how often to look: once a month is enough for most accounts, with a before-and-after check about four weeks after any major change. The component statuses compare you against other advertisers over a rolling 90 days, so they move gradually, and checking weekly tells you little beyond the fact that you checked. A high score will not lower your costs by itself, but the ad quality behind it will, which makes a monthly look at the Below average rows one of the better uses of twenty minutes in the account.
Vignesh runs paid media, social and design at Enigma Scale. He has managed spend across every major ad platform and rebuilt the sites those clicks land on, which is usually where the real problem turns out to be.
More from Vignesh →