TL;DR
Most Google Ads Quality Score advice is wrong, or at least incomplete. The visible 1-10 Quality Score isn’t the score used in the auction. Improving the visible number doesn’t always reduce CPC. And low Quality Score hurts in two ways most operators only count one of: higher cost per click AND your ads getting surfaced on lower-quality query matches. This article covers what Quality Score actually measures, why Google built it the way it did, the four-step diagnostic process for identifying which component is causing your low scores, and the specific patterns we see most often in 2026 Melbourne accounts.
The Quality Score misconception most operators still believe
The standard understanding of Quality Score looks like this: the score is a 1-to-10 number Google shows in the keyword report. Improving the number improves your ad position and lowers your CPC. Therefore, optimise the components, raise the number, win.
This understanding is incomplete in three ways that matter.
First, the visible 1-to-10 Quality Score is not the score used in the actual auction. Google’s documentation describes the visible QS as a diagnostic indicator. The real-time Quality Score used to calculate Ad Rank in each individual auction is calculated separately, on a finer scale, with more factors. The visible 1-to-10 is a simplified snapshot for diagnostic purposes, not the operational score that affects each auction.
Second, improving the visible QS doesn’t always reduce CPC. Smart Bidding strategies already factor real-time QS into their bid calculations. If you’re running a Smart Bidding strategy and you improve a keyword’s visible QS from 6 to 8, the bidding algorithm has often already been pricing in the QS factors. The visible score change confirms what the algorithm already knew. Your CPC may or may not move, depending on competitive dynamics and how strongly the algorithm was already compensating.
Third, low Quality Score hurts in ways that don’t show up in the CPC column at all. Which brings us to why Google built the system this way in the first place.
Why Quality Score exists, and what that tells you about how Google handles low scores
Google’s revenue model is paid per click. Higher click volume across the platform means higher revenue. This basic economic logic underpins how Google’s auction algorithms have evolved.
Within that logic, the algorithm strongly rewards ads that match user search intent. Matched ads get clicked at higher rates than unmatched ads, which means more revenue per impression. Quality Score is Google’s mechanism for incentivising advertisers to do the matching work themselves. Build ads that match search intent, build relevant landing pages, and Google rewards you with lower costs and better ad positions.
The mechanism works as intended when advertisers have good Quality Score. But there’s an underexamined corner of how the system behaves when they don’t.
When a keyword has low Quality Score, two things happen. The first is well-understood: the advertiser pays more per click for that keyword to maintain ad position. The second is less commonly discussed: Google’s algorithm has to maintain click volume across the platform somehow. If your low-QS ad isn’t getting clicks at the rate Google needs for its revenue model, the algorithm compensates by surfacing your ad on looser query matches. Broader, lower-intent queries where the ad-to-intent matching is poor but click probability is still adequate to maintain platform-level click rates.
The implication is the operator insight most QS advice misses. Low Quality Score doesn’t just inflate your CPC. It also spreads your spend across query matches you’d never knowingly target. You can see this pattern in search-term reports of accounts with multiple low-QS keywords: the search terms widen significantly, more loosely-matched traffic appears, and conversion quality degrades alongside the CPC inflation.
This is why low QS is a two-way cost. The visible cost is higher CPC. The hidden cost is worse query match quality. And the second cost often exceeds the first.
The operator framing this leads to: improving Quality Score does more than reduce CPC. It tightens which queries your ads appear for, which improves the entire account’s spend efficiency, not just the per-click cost.
What’s changed about Quality Score since 2022
Three shifts make Quality Score diagnostics different in 2026 than they were three years ago.
Smart Bidding dampens the visible QS impact. Three years ago, most operators ran Manual CPC or Enhanced CPC bidding, where Quality Score directly translated to CPC reduction in a way you could calculate predictably. With Smart Bidding now dominating, the algorithm has already factored real-time QS into its bid calculations. The relationship between visible QS improvement and observed CPC reduction has become less direct. This doesn’t make QS less important. It changes how you measure the impact of QS optimisation work.
Account-level QS history matters more than current state. Google’s algorithm tracks Quality Score patterns at the account level, not just at the keyword level. Accounts with persistent low Quality Score across multiple campaigns develop a reputation effect that’s harder to climb out of than a single low-QS keyword. This means newer accounts with clean QS history have an advantage that older accounts with persistent problems lack. The fix is methodical structural improvement, not isolated keyword tweaks.
Performance Max doesn’t show Quality Score, and that’s intentional. PMax campaigns have no visible Quality Score because the campaign type is built around the algorithm controlling all the match-quality decisions. Operators looking for “PMax Quality Score” miss the point. The Quality Score signal is being calculated and used, just not surfaced to operators. The 2026 implication: traditional Quality Score diagnostics apply primarily to Search and Shopping campaigns, not the increasing share of spend flowing through PMax. This shifts where QS optimisation work has highest impact.
The three components: what each actually measures
Quality Score has three visible components: Expected CTR, Ad Relevance, and Landing Page Experience. Each measures something specific, and the diagnostic value depends on understanding what each component actually evaluates.
Expected CTR
What Google measures: the predicted click-through rate for this ad on this keyword, adjusted for ad position. The “expected” framing matters because raw CTR isn’t comparable across positions (an ad at position 1 will always have higher raw CTR than the same ad at position 4). Expected CTR normalises for position, then compares to what’s typical for the keyword in question.
When Expected CTR is rated “below average,” it means your ad’s predicted click rate is lower than what’s typical for that keyword at the relevant position. Two common causes worth distinguishing. The first is genuine ad weakness: the headline doesn’t compel clicks compared to competitors. The second is keyword-ad mismatch: the keyword’s typical CTR profile is set by competitors targeting it with very different intent or messaging, and your ad doesn’t fit that profile.
The hidden factor most operators miss: match type plays heavily into Expected CTR. The same ad can show different Expected CTR ratings depending on whether the keyword is exact, phrase, or broad match, because the underlying search-term population differs. Broad match keywords often show lower Expected CTR ratings because the algorithm is calculating CTR against a wider, less consistent search-term distribution.
When low Expected CTR is a real problem versus when it’s misleading. Real problems: high-volume, well-defined intent keywords where competitors are getting better CTR with similar messaging. Misleading signals: broad match keywords on emerging or shifting topics, where the Expected CTR rating is partially an artifact of measurement noise.
Ad Relevance
What Google measures: the semantic matching between the keyword and the ad copy. The algorithm scans the ad’s headlines and descriptions, compares to the keyword and to other ads competing for the same keyword, and rates whether the ad clearly addresses the keyword’s intent.
Why “include the keyword in the headline” isn’t always enough. The algorithm has moved beyond simple keyword-presence checks. In 2024-2026, ad relevance scoring considers semantic similarity, not just literal keyword inclusion. An ad headline that contains the keyword but isn’t semantically connected to the searcher’s likely intent can still score below average. Conversely, an ad headline that doesn’t contain the literal keyword but is semantically very tight to the intent can score above average.
When ad relevance “below average” is a structural problem versus a quick-fix problem. Structural problems: a single ad group containing keywords that span multiple intents (commercial, informational, navigational mixed together). No single ad can be relevant to all of them. The fix is restructuring into single-theme ad groups, not rewriting the ad. Quick-fix problems: a tight ad group where the ad copy has drifted away from the keyword theme over time. The fix is rewriting the ad to reconnect with the keyword theme.
Landing Page Experience
What Google measures: not just page speed, though that matters. The algorithm evaluates relevance of the landing page content to the ad and keyword, mobile-friendliness, transparency about who you are and what you offer, and ease of navigation. Page speed is one component, not the whole signal.
Mobile-specific factors that matter more in 2026. The mobile experience now weighs more heavily than the desktop experience in landing page scoring, mirroring how users actually arrive at most landing pages. Mobile-specific issues: forms with too many fields, slow-loading hero images, intrusive interstitials, navigation that requires excessive tapping. These can drag down landing page experience even when the desktop version is fine.
Why landing page experience is often the last component to fix, not the first. Landing page fixes are usually the most expensive component to address. They require dev resources, design changes, content rewrites, and conversion rate testing. Ad relevance fixes are usually the cheapest (rewrite the ad). Expected CTR fixes are mid-tier (rewrite ad copy more substantially, possibly restructure ad groups). The cost-benefit order matters when deciding what to tackle first.
The trade-off worth naming. Page changes that improve Quality Score landing page experience can sometimes reduce conversion rate. A faster page with simpler content can have better landing page experience signals but convert worse if it strips out information visitors actually needed. The QS improvement isn’t worth a conversion rate decline. Test before generalising.
The 4-step diagnostic process
A systematic way to approach Quality Score that prioritises high-impact fixes over keyword-by-keyword tweaks.
Step 1: Filter for the keywords that actually matter
Quality Score data is noisy at low volumes. Apply two filters before diagnosing anything.
Filter by impressions threshold. Any keyword with fewer than 100 impressions in the last 30 days doesn’t have enough data for Quality Score to be reliable. The visible score might say 3 or 4, but the underlying calculation has so little input data that the rating is largely measurement noise.
Filter by business impact. Sort keywords by conversion volume or conversion value. Quality Score optimisation work on a keyword driving zero conversions is wasted effort, regardless of how low the score is. Focus first on keywords that are actually driving business outcomes (or that should be, based on commercial intent).
The counter-conventional point worth naming: not every low-QS keyword needs fixing. Some low-QS keywords are low-QS because they’re low-value, and the right action is to pause them, not to optimise them.
Step 2: Decompose the QS by component
View the three Quality Score columns in the keyword report (you’ll need to add them via the columns panel if they’re not visible). For each problem keyword, identify which of the three components is rated below average.
The diagnostic question shifts from “what’s the score” to “which component is failing.” Patterns matter more than individual scores. If ad relevance is below average on multiple ad groups, the issue is probably structural. If expected CTR is below average specifically on long-tail keywords, the issue is probably the ad copy not addressing the long-tail intent. If landing page experience is below average across the whole account, the issue is probably the landing pages, not the ads.
Step 3: Fix by component, not by keyword
Group the problems by which component is failing, then address each component systematically rather than keyword-by-keyword.
For ad relevance issues across multiple ad groups, the fix is usually structural. Restructure the ad groups to be more thematically tight. Single-theme ad groups (where every keyword shares a clear intent) consistently score better on ad relevance than mixed-theme ad groups, regardless of how well-written the ads are.
For expected CTR issues, the fix is usually copy work. Rewrite ad copy with stronger CTAs, more specific value propositions, and benefit-driven hooks. Test different headlines. Pay particular attention to whether the ad copy is differentiated from the other ads competing for the same keyword.
For landing page experience issues, the fix is technical and content work. Audit page speed (particularly mobile), check mobile usability, review content relevance to the ads pointing to the page. Often the right answer is creating dedicated landing pages for the highest-traffic ad groups rather than pointing all ads to generic service pages.
The order matters. Ad relevance fixes are usually cheapest and fastest. Expected CTR fixes are medium cost. Landing page fixes are usually most expensive. Address in order of cost-effectiveness, not just severity.
Step 4: Measure the right outcome
Don’t measure “did the Quality Score number go up.” That’s a lag indicator and not the actual business outcome you care about.
Measure the business outcomes that QS optimisation is meant to drive. CPC reduction is the obvious one, but as the Smart Bidding section noted, the relationship isn’t as direct as it used to be. Better metrics to track: search-term quality (are the actual queries that ads are matching to becoming more relevant), ad rank for high-intent keywords (are you appearing in better positions for the queries that matter), and conversion rate from search ads (because tighter query matches usually mean higher-intent traffic, which usually converts better).
The honest framing. A small Quality Score improvement on a high-volume, high-conversion keyword is more valuable than a large Quality Score improvement on a low-volume, low-conversion keyword. The metric that matters is total business impact, not aggregate QS movement.
When low Quality Score isn’t the problem to fix (self-qualification)
Conditions where time spent on Quality Score optimisation is wasted, and the right move is to do something else.
When Smart Bidding is running well and the account is profitable despite low visible QS. The algorithm has often already optimised around the visible QS issue. Spending operator time on QS improvements that won’t change the bid algorithm’s behaviour is wasted effort.
When conversion economics are strong despite low QS. If you’re paying more per click than you would with better QS but the math still works (cost per acquisition or return on ad spend meet targets), QS optimisation is a marginal lever, not a critical one. Other work has higher impact.
When the keyword is a brand keyword. Quality Score on brand terms is often a measurement artifact, not a real problem. Brand keywords have high CTR by definition (people searching for your brand intend to find you), and the QS rating sometimes fails to reflect this correctly. Focus elsewhere.
When the keyword is high-intent commercial and the alternative is not appearing at all. For a small number of high-conversion-intent keywords, even an inflated CPC is worth paying because the alternative is losing the lead entirely to competitors. QS optimisation on these keywords has value, but the urgency is lower than on broader keyword sets.
When there’s less than 30 days of data. Quality Score calculations need a meaningful data window to be reliable. Acting on QS signals before 30 days is acting on noise.
The honest framing. In 2026, Quality Score optimisation is one tool among many, not the primary lever it was in 2018. Operators who still treat it as the highest-impact work haven’t updated their methodology.
Specific Quality Score patterns we see most often in 2026
Three patterns that show up repeatedly in Adgenix audits of Melbourne accounts.
Pattern 1: Ad relevance “below average” across themed ad groups
Often the diagnostic surfaces ad relevance below-average across multiple ad groups that look themed but actually aren’t tight enough. The ad group might contain 15 keywords spanning slightly different intents (e.g., “lead generation services” mixed with “lead gen agency” mixed with “lead generation specialists”). Each keyword has slightly different intent, no single ad can be relevant to all of them, and the algorithm rates ad relevance below average for most of them.
The fix is structural. Split the ad group into 2-3 tighter ad groups, each with a single dominant intent. Ad relevance scores typically improve across most of the keywords within 30-60 days.
Pattern 2: Expected CTR collapse after AI-generated copy refresh
A pattern that’s emerged specifically in 2024-2026. Operators run Responsive Search Ad copy through AI tools without sufficient human editing, ending up with headlines that are technically grammatical but lack the concrete benefit-driven specificity that drives clicks. Expected CTR ratings drop across affected ad groups, often by 1-2 points.
The fix is human-edited copy. AI tools are useful for generating starting points and variants, but the headlines and descriptions need human review to ensure they include specific benefits, concrete numbers where possible, and the kind of differentiated messaging that AI-generated copy tends to flatten out of.
Pattern 3: Landing page experience flagged for accounts running both PMax and Search campaigns
When an account runs both PMax campaigns and traditional Search campaigns, often pointing to the same landing pages, the landing page experience signal can deteriorate for the Search campaign ads. PMax-driven traffic and Search-driven traffic have different page-engagement profiles, and pages optimised for PMax can underperform on Search Quality Score signals.
The fix is dedicated landing pages. Search campaigns should point to landing pages designed for Search-driven traffic profiles (more text-heavy, more comparison-focused, more trust signals). PMax campaigns can point to different pages designed for the broader-intent traffic that PMax delivers.
Honest CPC impact ranges
Quality Score improvements affect CPC at varying levels depending on starting point, bid strategy, and competitive dynamics. Specific ranges in any one account vary widely.
The typical pattern. Improvements from very low scores (3-5 range) tend to deliver the largest CPC reductions, often in the 30-50% range when the underlying components are genuinely broken. Improvements from moderate scores (6-7) deliver smaller gains, typically 10-25%. Improvements above QS 8 deliver diminishing returns, often not worth the operator effort relative to other work.
These ranges assume Manual CPC or Enhanced CPC bidding. With Smart Bidding strategies, the visible CPC impact is often smaller because the algorithm has already been compensating. The real benefit shows up in different metrics: better ad position for high-value keywords, improved query match quality, and reduced spend on poor-quality search terms.
Your account may differ substantially from these ranges. Competitive density, conversion economics, and bid strategy choice all affect what QS improvements actually deliver in practice. Use the ranges as rough guidance, not as commitments.
What Quality Score optimisation fixes (and doesn’t)
Quality Score optimisation is one tool. Knowing what it does and doesn’t address keeps the work proportionate.
What QS optimisation fixes. CPC inflation from low scores, particularly when bid strategy is Manual CPC. Auction visibility for high-intent keywords (ad rank improves with QS). Query match quality (better QS reduces the algorithm’s tendency to surface your ads on loose query matches). Account-level reputation effects over time, particularly for newer accounts establishing baseline performance patterns.
What QS optimisation doesn’t fix. Wrong conversion signals. If your account is optimising toward a conversion goal that doesn’t reflect business outcomes, QS improvements won’t help. The conversion tracking work needs to happen first.
Wrong account structure. If campaigns are organised around the wrong logic (by product category when they should be by margin tier, or by location when they should be by buyer intent), QS improvements paper over a structural problem that needs direct attention.
Wrong campaign type for the use case. Putting your full e-commerce inventory into a single Performance Max campaign when some products need Standard Shopping treatment is a structural choice that QS work won’t fix.
Smart Bidding accounts already optimising around the issue. The algorithm has often priced QS into bids already. QS improvements may show in ranks and query matches but not in CPC.
The diagnostic order matters. Fix conversion tracking first (using the systematic negative keyword strategy and related foundation work). Fix account structure second (using the 10-section audit framework to identify structural issues). Fix Quality Score third, once the foundation is sound.
If your Quality Score audit suggests structural issues
For senior operators looking at Quality Score across their account, the pattern that emerges from the diagnostic above usually points to one of three issues: structure that’s too broad (Pattern 1), ad copy that has drifted from intent (Pattern 2), or landing page misalignment (Pattern 3). Each has a different fix and a different time investment.
If you’re evaluating whether to take this on yourself or hire a senior Google Ads specialist in Melbourne, the Free Revenue Leak Report includes a Quality Score component diagnostic as part of the 48-hour audit. The report identifies which of the three patterns is affecting your account and what the rough scope of fixing each would look like.
If you’re earlier in the decision and not sure whether freelance or agency is the right model for the work involved, the freelancer vs agency decision framework covers that question first.
For everything else, the diagnostic order in this article applies regardless of who ends up doing the work.
The one diagnostic question that surfaces real Quality Score issues
If you take only one thing from this article into your own account, take this diagnostic question:
“If I look at my top 20 keywords by spend and filter to only those below Quality Score 7, what percentage of my monthly spend is currently flowing through low-QS keywords?”
The answer reveals whether QS optimisation work is genuinely high-impact for your account or not.
If the answer is more than 30%, Quality Score is materially affecting your account economics. The optimisation work is worth doing, in the order outlined in this article (component-level diagnosis, structural fixes first, copy fixes second, landing page fixes third).
If the answer is between 10% and 30%, QS optimisation is one of several worthwhile levers but not the highest priority. Other work (conversion tracking quality, account structure, bid strategy fit) likely has higher impact. For accounts in this range, working with a senior Google Ads consultant in Melbourne on a broader account audit usually produces better outcomes than focusing on Quality Score in isolation.
If the answer is less than 10%, your QS profile is already strong enough that further optimisation hits diminishing returns. Focus operator effort elsewhere.
Most operators have never run this calculation on their own accounts. They know their account-level QS average or remember a few specific keywords with low scores, but they’ve never quantified what percentage of their monthly spend is flowing through low-QS territory. The number is usually different from what they’d estimate.
Run the calculation. Then decide whether Quality Score deserves the operator hours, or whether other work has higher impact right now.
Adgenix · Senior Paid Media · Melbourne, Australia
