Most advice on how to improve Quality Score in Google Ads stops at “write more relevant ads” and “make your landing page faster”. Useful, but useless when you are staring at 400 keywords and no idea which fix applies to which keyword. There is a practical rundown of it online.
Quality Score is not one number you push up. It is three separate diagnostics stacked into a single 1 to 10 value, and each one is fixed by a different action. Pull the wrong lever and nothing changes for weeks.
This guide breaks down the three components, shows you exactly which lever to pull based on which component is rated Below Average, how to surface the right columns in the interface, and what score ranges realistically mean for your cost per click.
What Quality Score actually is (and what it is not)
Quality Score is a keyword-level diagnostic on a 1 to 10 scale, available in Search campaigns. It is Google’s retrospective estimate of how your keyword, ads and landing page compare with other advertisers competing on the same terms.
Three things you need to be clear on before you start optimising:
- It is a diagnostic, not the auction input. The live auction uses real-time quality signals calculated per query, device, location and context. Quality Score is the aggregated summary you get to look at afterwards.
- It is calculated from exact-match search activity. Impressions where your keyword matched loosely still cost you money, but the score itself is anchored on exact matches of that keyword.
- A score of “null” (shown as a dash) is not bad. It means not enough exact-match impression data yet.
The three components, in plain terms
| Component | What Google is measuring | What you control |
|---|---|---|
| Expected CTR | How likely your ad is to be clicked when it shows for that keyword, normalised for ad position | Ad copy appeal, offer, assets, historical click performance, query quality |
| Ad relevance | How closely your ad message matches the intent behind the keyword | Ad group structure, keyword-to-headline mapping, wording overlap |
| Landing page experience | How useful and easy the destination page is for someone arriving from that keyword | Message match, page content, speed, mobile usability, transparency, navigation |
Each component is rated Below Average, Average or Above Average. That wording is the whole game: it is relative to other advertisers on the same keywords over the same period, not an absolute grade of your work.
Step 1: Pull the Quality Score columns (2 minutes)
You cannot diagnose anything from the single 1 to 10 number. Add the component columns:
- Open Google Ads and go to Campaigns > Audiences, keywords and content > Search keywords.
- Click the columns icon above the table, then Modify columns.
- Open the Quality Score section.
- Tick: Quality Score, Exp. CTR, Ad relevance, Landing page exp.
- Also tick the historical versions: Qual. Score (hist.), Landing page exp. (hist.), Ad relevance (hist.), Exp. CTR (hist.).
- Save the set as a layout called something like “QS Audit” so you can reload it in one click.
Why the historical columns matter: the standard columns only show the current status. Historical columns show the value as it was during the date range you selected, which is the only way to prove that a change you shipped six weeks ago actually worked. The reasoning is set out in this piece.
Two more things worth doing while you are in there:
- Sort by Cost descending, not by Quality Score. A 4/10 keyword spending 12 euros a month is noise. A 6/10 keyword spending 3,000 euros is your priority.
- Segment by Device. Landing page experience problems very often live only on mobile.
Step 2: The diagnostic table (which lever to pull)
Find the pattern that matches your keyword, then work the corresponding lever. Do not work all three at once, you will never know what moved the number.
| Component rated Below Average | Most likely root cause | Lever to pull | Typical time to see movement |
|---|---|---|---|
| Ad relevance only | Keyword sits in an ad group covering too many intents | Lever 1 and Lever 2 (restructure, then rewrite headlines around the keyword theme) | 1 to 3 weeks |
| Expected CTR only | Ad is relevant but unappealing, or the keyword pulls in poor-quality matched traffic | Lever 3 and Lever 4 (offer-led copy, assets, search term hygiene) | 2 to 6 weeks |
| Landing page experience only | Traffic lands on a generic page, or the page is slow or thin on mobile | Lever 5 and Lever 6 (message match, speed and usability) | 3 to 8 weeks |
| Ad relevance + Expected CTR | Ad group theme is broken; the ad speaks to nobody in particular | Split the ad group first, then rewrite. Restructure beats copy tweaks here | 2 to 4 weeks |
| Ad relevance + Landing page exp. | You are bidding on a term your site does not genuinely serve | Lever 7: build the missing page, or drop the keyword | 4 weeks or more |
| All three Below Average | Wrong keyword for this business, or a competitor set you cannot win cheaply | Lever 7: pause, or isolate into a dedicated campaign with its own page and budget | Decision, not an optimisation |
| All three Average, score stuck at 5 or 6 | Nothing broken, nothing outstanding | Lever 4 (assets and extensions) gives the fastest incremental lift | 2 to 4 weeks |
The 7 levers that actually move the needle
Lever 1: Rebuild ad groups around single intents, not single keywords
The old advice was one keyword per ad group. With broad match, smart bidding and responsive search ads, that creates data fragmentation and hurts you. What still works is one intent per ad group.
- Group by what the searcher wants, not by string similarity. “content agency pricing”, “how much does a content agency cost” and “content agency rates” belong together. “content agency london” does not.
- Aim for ad groups where a single set of three headlines could honestly answer every keyword inside it. If it cannot, split.
- Keep 5 to 20 keywords per ad group as a working range, adjusted for volume.
Ad relevance responds to structure faster than to anything else you can do.
Lever 2: Get the keyword theme into the headlines that actually serve
With responsive search ads, Google mixes assets, so “I put the keyword in the ad” is no longer enough. It has to be in the assets that get served.
- Write at least 4 of your 15 headlines around the exact keyword theme, phrased differently each time.
- Check the asset details report and look at asset performance ratings. Replace anything marked Low. Cull, do not accumulate.
- Use pinning sparingly: pin one keyword-carrying headline to position 1 when relevance is Below Average, leave positions 2 and 3 open so Google can still optimise. Pinning everything usually costs you more in expected CTR than it gains in relevance.
- Put the keyword theme in Path 1 of the display URL. It is a small relevance and CTR signal that costs nothing.
- Repeat the theme in at least one description, then spend the rest of the space on the offer.
Lever 3: Cut the matched traffic that drags expected CTR down
Expected CTR is calculated on exact matches, but sloppy matching still trains a weak click profile for the keyword and burns budget. Weekly hygiene:
- Open the Search terms report for the last 14 days.
- Filter for clicks > 0, conversions = 0 and sort by cost.
- Add negatives for anything informational when you sell, or commercial when you educate. Job seekers, DIY, free, competitor names you do not want, and student research queries are the usual suspects.
- Build shared negative keyword lists by theme (jobs, free, DIY, competitors) and apply them account-wide instead of rebuilding per campaign.
- Add strong performing search terms as their own keywords, in the ad group whose ads already match them.
Lever 4: Fill every asset slot, because bigger ads get clicked more
Expected CTR is position-normalised, but ad real estate is not. Ads with more assets occupy more space and earn more clicks at the same rank. Minimum viable set per campaign or ad group:
- 4 sitelinks minimum (aim for 6 to 8), with descriptions filled in
- 4 callouts covering proof points, not slogans
- 2 structured snippets with relevant header types
- Image assets and a logo, since these strongly influence CTR on mobile
- Promotion, price and lead form assets where they apply
- Call and location assets if the phone or the shop matters to your model
This is the highest ratio of effort to result on the whole list, and it is the one most accounts skip.
Lever 5: Fix message match on the landing page before touching design
Landing page experience is not a beauty contest. Google is judging whether someone who clicked that specific ad for that specific keyword got what they came for. Non-negotiables:
- The H1 echoes the ad headline and the keyword theme. If your ad says “B2B SaaS content writing”, the page should not say “Digital solutions for modern brands”.
- The offer and the primary call to action are visible above the fold on mobile, without scrolling.
- The page contains original, substantial content on the topic: specifics, deliverables, pricing signals, proof.
- Trust and transparency are present: contact details, a real address, clear terms, privacy policy, refund or delivery information for ecommerce.
- Form friction is minimal. Every optional field you delete helps conversion and, indirectly, the engagement signals behind this component.
- Navigation still works. Fully stripped pages with no way out can read as low quality.
Lever 6: Treat page speed and mobile usability as a QS lever, not a dev backlog item
Speed is an explicit part of landing page experience, and mobile is where most Search clicks happen. Practical targets:
- Largest Contentful Paint under 2.5 seconds on mobile 4G
- Interaction to Next Paint under 200 ms
- Cumulative Layout Shift under 0.1 (banners and late-loading fonts are the usual culprits)
Quick wins that rarely need a redesign: compress and lazy-load images below the fold, serve modern image formats, defer non-critical scripts, audit your tag manager for tags nobody uses any more, remove chat widgets that block the main thread, and cache aggressively. Also verify your final URLs for redirect chains and soft 404s, since a keyword pointing at a redirecting or dead URL will quietly tank this component.
Lever 7: Prune or isolate the keywords that will never recover
Some keywords are structurally bad for your account. A generic head term that competes with marketplaces will keep a Below Average expected CTR no matter how good your copy is. Decision framework:
- Converting profitably at 4/10? Keep it. Quality Score is a cost efficiency indicator, not a KPI. A profitable 4 beats an unprofitable 9.
- Expensive, low QS, no conversions after enough clicks? Pause it and reinvest in the terms that already work.
- Strategically important but low QS? Isolate it in its own ad group with its own dedicated landing page and its own ads. Isolation is the only reliable way to give a stubborn keyword a fair chance.
What Quality Score ranges realistically mean for your CPC
Google does not publish a discount table, and anyone who shows you one as official is guessing. What is well documented is the mechanism: Ad Rank combines your bid with quality signals, so higher quality means you can hold the same position at a lower cost per click, and you may qualify for ad positions and assets you would otherwise be priced out of.
The table below is a directional model based on the widely observed inverse relationship between quality and CPC, using a score of 5 as the neutral baseline. Treat it as an order of magnitude, not a promise.
| Quality Score | Directional CPC effect vs a score of 5 | What it usually means in practice |
|---|---|---|
| 9 to 10 | Roughly 40% to 50% cheaper | All three components Above Average. Rare, and usually on branded or tightly niched terms |
| 7 to 8 | Roughly 25% to 37% cheaper | The realistic target zone for commercial non-brand keywords. Stop optimising here and go work on conversion rate |
| 6 | Roughly 15% cheaper | Usually two Average and one Above Average. One lever away from 7 or 8 |
| 5 | Baseline | All three Average. Nothing broken, nothing exceptional |
| 3 to 4 | Roughly 25% to 65% more expensive | At least one component Below Average. Highest return on your optimisation time |
| 1 to 2 | Multiples of the baseline, plus impression loss | Usually a structural mismatch. Fix the landing page or drop the keyword |
What is a good Quality Score? A weighted average of 7 or above on your non-brand keywords is a healthy account. Branded terms commonly sit at 8 to 10 and inflate the picture, so always calculate your average excluding brand, and weight it by cost rather than by keyword count.
A 30-day working cadence
- Day 1: Load your QS Audit column layout. Export keywords with cost above your threshold, sorted by spend. Tag each one with the failing component.
- Days 2 to 5: Ad relevance work. Split over-broad ad groups, rewrite headlines for the top 10 spenders. One change per ad group.
- Days 6 to 10: Expected CTR work. Complete asset coverage, refresh Low-rated assets, add negatives from the search terms report.
- Days 11 to 20: Landing page work. Fix message match on the top 3 destination pages, then attack LCP and mobile layout.
- Day 21: Search terms review again. Keep the negative lists moving.
- Day 30: Compare historical columns period over period. Keep what moved, revert what did not, and pick the next batch by spend.
Three habits that quietly kill Quality Score progress: changing all three components in the same week so you learn nothing, judging results after 5 days when the data window is far longer, and chasing 10/10 on keywords that are already profitable instead of fixing the expensive 4s. Improve Your Performance with Google Ads Quality Score is a useful companion to this.
FAQ
Where do I see Quality Score in Google Ads?
Go to Campaigns > Audiences, keywords and content > Search keywords, click the columns icon, choose Modify columns, open the Quality Score section and add Quality Score plus the three component columns and their historical versions. It exists for Search keywords only.
Why does my keyword show a dash instead of a score?
Not enough exact-match impressions yet. Google needs meaningful exact-match volume before it reports a score. Very new keywords, very low volume keywords and paused keywords often show a dash indefinitely.
How long does it take to improve Quality Score?
Ad relevance can shift within 1 to 3 weeks of a restructure. Expected CTR usually takes 2 to 6 weeks because it needs new click data to accumulate. Landing page experience is the slowest, often 3 to 8 weeks after the page changes are live and re-crawled.
Is Quality Score used in Performance Max or Demand Gen?
No keyword-level Quality Score is reported there. The same underlying quality principles still apply though: asset relevance, landing page experience and speed all influence how those campaigns perform. In those campaign types you work with Ad Strength and asset performance ratings instead.
Does raising my bid improve Quality Score?
Not directly. Bids are a separate Ad Rank input. Higher bids can win better positions, which brings more impressions and more click data, which can indirectly speed up how quickly a component recalibrates. Buying your way to a better score is not a strategy.
Can I see a competitor’s Quality Score?
No. There is no public Quality Score checker for other advertisers, and any tool claiming to reveal it is estimating. Use Auction insights and impression share metrics to infer competitive strength instead.
Should I aim for 10/10 on every keyword?
No. Quality Score is a means to cheaper clicks, not the goal. Once a keyword sits at 7 or 8 and converts profitably, your time is better spent on conversion rate, offer strength and budget allocation.
Bringing it together
Improving Quality Score in Google Ads is a diagnostic exercise, not a checklist. Pull the component columns, sort by spend, identify which of the three ratings is Below Average, then pull the one lever that matches. Give it a real data window, verify it with the historical columns, and move to the next batch.
Do that consistently and the score follows the work, along with the lower CPCs, the better positions and the extra clicks you were previously paying a premium to lose.
Need help fixing the landing page and ad copy side of the equation? That is exactly the kind of work we do at King Content Agency. Get in touch and we will audit your top-spending keywords first.