Google Ads Quality Score: What It Actually Measures, and How to Improve It in 2026

Somewhere in every Google Ads account there’s a column most people stopped looking at years ago. It sits quietly next to your keywords, showing a number from 1 to 10, and unless you’ve gone out of your way to add it to your view, you probably haven’t seen it in months. That column is Quality Score, and the story you’ve likely heard about it goes something like this: it used to matter, then Smart Bidding and automation took over, and now it’s a relic — a vanity metric nobody needs to think about anymore.

That story is only half true, and the half that’s wrong is costing accounts real money. Quality Score itself is diagnostic — Google has always been clear that it’s a report card, not a lever you pull directly. But the three things it measures are not diagnostic at all. They are living inputs into the real-time auction that decides whether your ad shows and what you pay when it does, whether you’re bidding manually, running Target ROAS, or letting Performance Max make every decision for you. Ignore what Quality Score is telling you, and you’re not opting out of a metric — you’re opting out of understanding why your cost per click is what it is.

This post is a practical walkthrough of what Quality Score actually measures, why the “–” shows up where a score should be, how it still feeds into Ad Rank under automated bidding, and what to actually do about each of its three components. None of this is about chasing a number for its own sake. It’s about understanding the mechanism well enough that when your CPCs creep up or your impression share won’t budge, you know where to look.

What Quality Score Actually Measures (and What It Doesn’t)

Quality Score is a keyword-level rating on a 1-to-10 scale that Google Ads assigns to estimate how relevant your keyword, ad, and landing page are likely to be to someone searching that term. It isn’t visible by default — you have to add the Quality Score column (along with its three component columns) to your Keywords view to see it at all, which is probably why so many accounts have quietly stopped tracking it.

The important nuance, and the one most explanations skip, is that the number you see in that column is a historical, diagnostic estimate. It’s Google’s way of summarizing, at a glance, how your keyword has performed relative to competitors bidding on the same terms. It is not the exact number plugged into the auction the next time someone searches. The live auction uses a real-time version of the same underlying signals — expected clickthrough rate, ad relevance, and landing page experience — combined with context at that specific moment (device, location, time of day, the exact search query, and more) to calculate Ad Rank. Quality Score is the closest thing you get to a stable, reportable proxy for those real-time signals, which is exactly why it’s still worth reading even though it isn’t literally what determines each auction.

Two things follow from that distinction. First, don’t panic over small week-to-week swings in a single keyword’s score — the reported number is an aggregate, and the real-time signals it’s approximating move constantly. Second, don’t dismiss it either, because a keyword that sits at 3/10 for months isn’t a fluke; it’s telling you something durable about how Google’s systems perceive your ad and page relative to everyone else competing for that query.

The Three Components Behind Every Score

Quality Score is built from three status ratings, each shown as “Below average,” “Average,” or “Above average” rather than a raw number:

• Expected click-through rate (CTR) — how likely people are to click your ad when it’s shown for that keyword, compared to other advertisers whose ads appeared for the same search terms. This is judged independently of your actual ad position, so a lower position doesn’t automatically drag this rating down.
• Ad relevance — how closely your ad’s message actually matches the intent behind the keyword. This is where mismatched match types and loosely themed ad groups do the most damage: if your ad group contains keywords with meaningfully different intent, no single ad can be relevant to all of them at once.
• Landing page experience — how relevant, transparent, and easy to navigate your landing page is for someone who just clicked that specific ad. Google evaluates this per keyword, not just per URL, because the same page can be a great landing experience for one search and a poor one for another.

Each of these three ratings is assessed relative to other advertisers competing for the same keywords, not against some fixed universal bar. That’s a meaningful detail: your landing page experience rating can drop not because your page got worse, but because competitors’ pages got better. It also means Quality Score is a moving target by design — you can do nothing and still watch a component rating slide, simply because the competitive bar around you shifted.

The overall 1-10 score is a rough composite of these three ratings, weighted so that landing page experience and ad relevance tend to carry meaningful weight alongside expected CTR — which is part of why “just write a better headline” is rarely a complete fix on its own. A keyword generally needs to be solid across all three, not brilliant on one and neglected on the other two, to sit in the 7-10 range that’s considered healthy.

Why You Sometimes See “–” Instead of a Score

If you’ve opened the Quality Score column and seen a dash instead of a number on some keywords, that isn’t a bug. Google needs a minimum amount of recent search volume and impression data on a keyword before it has enough signal to produce a reliable rating. Low-search-volume keywords — the long-tail, highly specific terms that individually get very few impressions — often sit at “–” indefinitely, simply because Google never accumulates enough data on that exact keyword to compare it against competitors with any confidence.

This matters for how you read your account. A blank Quality Score isn’t a red flag, and chasing a rating for a keyword that gets three impressions a month is a waste of effort. It’s a symptom of an account structure decision (how granular your keywords are), not evidence that something is broken. If a keyword you actually care about is consistently blank, the more useful question is usually whether it’s worth its own line item at all, or whether it should be folded into a broader match type or theme where Google has enough volume to actually judge relevance.

How Quality Score Feeds Into Ad Rank and Your Actual CPC

The reason any of this is worth your time is the direct line between Quality Score’s components and what you pay. Ad Rank — the score that determines both whether your ad shows and what position it lands in — is calculated from your bid combined with the same quality signals Quality Score reports on, plus the expected impact of ad extensions and formats, and context specific to that auction.

The practical consequence is one Google has stated plainly: a keyword with a higher Quality Score can achieve the same ad position at a lower cost per click than a competitor with a lower score and a higher bid. Two advertisers can bid the same amount and land in very different positions — or bid different amounts and land in the same position — purely because their underlying quality signals differ. This is why two accounts targeting identical keywords in the same market can have meaningfully different CPCs even when their bidding looks similar on paper. It’s also why “just raise the bid” is often the most expensive way to fix a position problem that’s actually a relevance problem.

This is the piece that gets lost when people write Quality Score off as obsolete: even if you never look at the 1-10 number again, the auction is still scoring the same three things about your ads and pages, every single time, and pricing your clicks accordingly.

Does Quality Score Still Matter With Smart Bidding or Performance Max?

This is the honest source of the “Quality Score doesn’t matter anymore” narrative, and it deserves a straight answer. If you’re running Target CPA, Target ROAS, Maximize Conversions, or Performance Max, you’re not manually setting bids per keyword — Google’s bidding systems are making that call in real time based on predicted conversion likelihood and value. That’s a genuine shift in where control sits, and it’s worth understanding on its own terms; we’ve written separately about how to actually choose between Target CPA, Target ROAS, and Maximize Conversions, because that decision shapes how aggressively the algorithm will bid on your behalf.

But automated bidding doesn’t remove quality signals from the equation — it changes who’s pulling the lever. The bidding algorithm is still bidding into an auction where Ad Rank depends partly on expected CTR, ad relevance, and landing page experience. A keyword with weak underlying quality signals still costs more to win the same position; Smart Bidding just absorbs that cost into its predictions rather than surfacing it to you as a visible bid decision. You lose the granular, keyword-level feedback loop, but you don’t lose the underlying mechanics.

Performance Max makes this more opaque still, because it isn’t organized around keywords at all, so there’s no per-keyword Quality Score to check in the first place. That doesn’t mean quality stops mattering inside Performance Max — asset relevance, landing page quality, and message match are still doing real work in how the campaign’s automated targeting performs — it just means you’ve lost the diagnostic report card and have to infer quality issues from downstream metrics like conversion rate and asset-level performance ratings instead. If you’ve ever tried to work out why a Performance Max campaign is underperforming without being able to see inside it, that opacity is exactly the same problem in a different shape.

The practical takeaway: automation changes how visible quality problems are, not whether they exist. If anything, the loss of keyword-level Quality Score reporting under automated bidding and Performance Max is a reason to pay more attention to the components you can still influence — your ads, your account structure, and your landing pages — not less.

Improving Expected CTR: Write Ads People Actually Click

Expected CTR is graded relative to other advertisers on the same keywords, so the bar isn’t “is this a good ad” in the abstract — it’s “would someone searching this exact term click this ad over the alternatives currently showing.” A few things reliably move this component:

• Match the query, not just the theme. An ad that speaks directly to the specific problem behind a search (“same-day emergency plumber”) will outperform a generic ad (“professional plumbing services”) even if both are technically relevant, because expected CTR rewards specificity.
• Use responsive search ads properly, not lazily. Feeding an RSA a dozen near-identical headline variations doesn’t give Google’s testing engine much to differentiate. Genuinely distinct headlines — different angles, different proof points, different calls to action — give the system more to learn from and more room to find a combination that beats the competitive set.
• Lean on ad extensions and assets. Sitelinks, callouts, structured snippets, and other assets increase the visual footprint and expected engagement of your ad, and Google explicitly factors expected impact from formats into the auction, separate from the base ad copy.
• Prune ad groups where one ad is trying to serve too many intents. If your best-performing headline can’t be honest about what the landing page actually offers for every keyword in that ad group, the ad group is too broad — not the ad.

Expected CTR is graded independent of your actual position, which means you can’t blame a low rating purely on being in position four instead of position one. It’s a genuine signal about ad quality, not a proxy for how much you’re bidding.

Improving Ad Relevance: Tighter Themes, Fewer Surprises

Ad relevance breaks down for a predictable reason: an ad group holding keywords with meaningfully different intent, forced to share one or two ads that can’t honestly speak to all of them. The fix is rarely “write a better ad” — it’s usually restructuring which keywords sit together in the first place.

This is also where your search term report becomes a Quality Score tool, not just a wasted-spend tool. Every irrelevant query your keywords are matching to is a small, repeated hit to ad relevance and expected CTR at the same time, because Google is watching what people actually search when your ad shows, not just the keyword you bid on. We covered the mechanics of this in detail in how to read a Google Ads search term report, and the short version for Quality Score purposes is: a clean search term report and a healthy ad relevance rating tend to move together, because they’re both downstream of the same underlying problem — keywords catching traffic they were never meant to catch.

Broad match deserves a specific mention here, because it’s the match type most likely to quietly erode ad relevance without you noticing. Broad match can perform well when it’s paired with strong conversion tracking, tight negative keyword lists, and enough conversion volume for Smart Bidding to learn from — but the moment any of those three are missing, broad match tends to drift toward looser and looser matches over time, and every one of those loose matches is a query your ad wasn’t really built to answer. If you’re running broad match without a disciplined negative keyword process behind it, ad relevance is usually the first component to suffer, well before you notice it in your conversion numbers.

Improving Landing Page Experience: What Google Is Actually Grading in 2026

Landing page experience is the component most people either ignore entirely or reduce to “make it load fast.” Speed matters, but it’s one input among several, and Google’s own guidance points to page speed measured against Core Web Vitals thresholds: Largest Contentful Paint under 2.5 seconds, Interaction to Next Paint under 200 milliseconds, and Cumulative Layout Shift under 0.1. If your landing pages are missing those marks, especially on mobile, that’s a concrete, fixable place to start.

But speed alone doesn’t explain most below-average ratings. Google’s evaluation increasingly weighs whether the page actually delivers what the ad promised — message match between your ad copy and your headline, whether the specific product or service mentioned in the ad is easy to find on the page rather than buried three clicks deep, and whether navigation is clear enough that a visitor isn’t left guessing where they landed or why. Reporting on how this component is evaluated has pointed specifically to pages that send visitors to unexpected destinations or bury them in confusing navigation as a recurring driver of below-average ratings — not just slow pages, but pages that technically load fine and still leave the visitor unsure they’re in the right place.

A practical landing page checklist for Quality Score purposes:

• Does the page headline restate the specific offer from the ad, not just the general category?
• Can a mobile visitor find the actual call to action without scrolling past unrelated content first?
• Does the page load acceptably against Core Web Vitals thresholds on a real mobile connection, not just on your office wifi?
• Is there a clear, single next step, rather than a page competing for attention with unrelated navigation, pop-ups, or offers?
• If the ad promised a specific price, promotion, or product, is that exact thing visible without requiring a click to find it?

None of this requires a redesign. Most landing page experience problems are fixable with a dedicated page per campaign theme rather than routing every ad to the same generic homepage or category page — a homepage built for browsing is rarely a good landing page for someone who just clicked a specific, high-intent ad.

Worth noting separately: mobile and desktop landing page experience are evaluated as distinct experiences, not one blended score. A page that was designed and tested on a desktop monitor and never properly checked on an actual phone can carry a below-average rating for the majority of your traffic while looking perfectly fine every time someone on your team reviews it at a desk. If most of your clicks are coming from mobile devices, that’s the version of the page worth auditing first, not the one your team happens to look at most often.

A Practical Quality Score Audit You Can Run This Week

You don’t need to overhaul the account to get value out of this. A focused audit looks like:

• Add the Quality Score columns (Quality Score, Expected CTR, Ad Relevance, Landing Page Experience) to your Keywords view if they aren’t already there.
• Sort by Quality Score, ascending, and filter to keywords with meaningful spend — ignore the low-volume “–” keywords for now.
• Group your lowest scorers by which component is dragging them down. A cluster of keywords all weak on landing page experience points to a page problem; a cluster weak on ad relevance points to an ad group structure problem; a cluster weak on expected CTR points to an ad copy problem.
• Cross-reference against spend, not just score. A 3/10 keyword getting no traffic is a curiosity. A 3/10 keyword burning a meaningful share of budget is where the fix pays for itself fastest.
• Fix the structural cause, not the symptom. If three keywords share a low ad relevance rating because they’re crammed into one ad group with a mismatched ad, split the ad group before you touch the ad copy.

This isn’t a one-time project. Component ratings are relative to competitors, which means they can drift even when you change nothing — a competitor improving their landing page can quietly push your relative rating down without you doing anything wrong. That’s really the core argument for treating this as an ongoing check rather than an annual cleanup, and it connects to a broader question worth asking about your account generally: how often should you actually check your Google Ads account? Quality Score is one of the clearer examples of a metric where the answer is “more often than you think,” because the goalposts move even when you’re standing still.

Why Quality Score Is a Moving Target, Not a One-Time Fix

The theme running through all of this is that Quality Score isn’t something you fix once and forget. It’s a relative, rolling comparison against a competitive set that’s also changing, evaluated across three components that each respond to different levers, and it feeds into an auction that runs fresh every single time someone searches. An account that looks healthy on a Tuesday can have a below-average landing page rating by the following month simply because the competitive bar moved, or because a new ad group launched without the same care as the rest of the account.

This is part of why continuous account monitoring tends to catch quality problems earlier than periodic manual reviews do — a keyword sliding from average to below-average on landing page experience is a small, early signal, and it’s much cheaper to act on early than to discover three months later as a broader CPC creep you can’t immediately explain. This is the kind of drift Growera’s daily account monitoring is built to catch: checking the account every day rather than whenever someone remembers to log in, so a component rating sliding the wrong way gets flagged while it’s still a small, cheap fix rather than a pattern you have to reverse-engineer weeks later. If you’re weighing whether ongoing monitoring is worth it against doing these audits manually, it’s worth seeing what that actually costs before assuming a manual quarterly review is the cheaper option — the manual version just hides its cost in the CPC creep you didn’t catch in time.

Quality Score Questions Worth Answering Directly

What counts as a good Quality Score? As a rough guide, scores of 7-10 are generally considered healthy, 5-6 are middling and worth investigating, and anything from 1-4 usually points to a real problem in at least one of the three components rather than random noise. Context matters more than the raw number, though — a competitive, high-intent commercial keyword sitting at 6 might be perfectly normal for that category, while the same score on a keyword with little competition is a clearer red flag.

Can a low Quality Score stop your ads from showing at all? Indirectly, yes. Quality Score doesn’t have a hard cutoff, but because it feeds into Ad Rank, a sufficiently weak set of component ratings combined with a modest bid can push your Ad Rank below the threshold needed to enter the auction at all, especially for competitive keywords. In practice, this shows up as a keyword with impressions quietly drying up rather than a visible rejection.

How quickly can Quality Score improve after you fix something? There’s no fixed timeline, and that’s a direct consequence of how the components are measured — as a rolling comparison against recent activity, not a single audit that resets the moment you make a change. A meaningfully better ad or landing page typically shows movement over the following weeks as Google accumulates enough fresh performance data to update its estimate, not overnight. Expect gradual movement from consistent improvement, not an instant jump after one edit.

Does Quality Score apply to Shopping, Display, or Performance Max campaigns? Quality Score as the familiar 1-10, keyword-level rating is specific to Search campaigns. Shopping and Display don’t report it the same way, and Performance Max doesn’t expose it at all because it isn’t structured around keywords. That doesn’t mean quality stops being evaluated in those campaign types — relevance between your creative, your product data, and your landing page still influences performance — it just means you lose the direct, reportable number and have to judge quality indirectly through metrics like conversion rate and asset performance ratings.

Is it worth manually pausing every keyword under a certain score? Rarely, and definitely not as a blanket rule. A low score on a keyword with meaningful spend and poor conversion performance is worth acting on. A low score on a keyword that converts well despite the rating is a case where the keyword is quietly proving Quality Score wrong for your specific business, and pausing it on principle would cut a working keyword for the sake of a diagnostic number. Use the score to prioritize where to look, not as an automatic pause rule.

The Bottom Line

Quality Score itself is a diagnostic number, not a lever — but the three things it measures, expected CTR, ad relevance, and landing page experience, are live inputs into every auction your ads compete in, regardless of whether you’re bidding manually or handing the decision to Smart Bidding or Performance Max. Automation changed who’s pulling the bidding lever; it didn’t remove quality from the equation, it just made the feedback louder in some places (keyword-level scores) and quieter in others (Performance Max’s lack of keyword-level reporting).

The practical version of all this is simple, even if the mechanics underneath it aren’t: stop treating the Quality Score column as irrelevant just because you can’t directly control the number. Use it as a map to where your ads and pages are underperforming relative to competitors, fix the structural causes rather than the symptoms, and check back regularly, because the competitive bar it’s measured against never holds still. Do that, and the CPC benefit takes care of itself — not because you chased a score, but because you fixed the things the score was actually describing.

None of this requires treating Quality Score as the goal. It’s a symptom tracker, not a target. The accounts that get the most value from it are the ones that use it to ask better questions about their ads, their structure, and their pages — and then keep asking those questions on a regular cadence, rather than once a quarter when CPCs have already climbed far enough to notice.