Google Ads Quality Score: Why Sorting Is the Wrong Approach

In Google Ads you can sort keywords by Quality Score ascending and start with the 1s. That’s exactly what most guides on the topic tell you to do. It’s still the wrong approach, and the reason a lot of accounts spend months “optimizing QS” without seeing any real CPC improvement.

What Keyword Quality Score Actually Is

Keyword Quality Score is a diagnostic on a scale of 1 to 10, shown per search keyword in your Google Ads account. It’s made up of three components, each rated as “above average”, “average” or “below average”:

Component	What it measures
Ad relevance	How closely your ad matches search intent
Expected CTR	How likely users are to click on your ad
Landing page experience	How useful and relevant your destination page is for the search

Hal Varian, former Chief Economist at Google, has said publicly that expected CTR drives up to 60% of the score (Google talk on YouTube). Ad relevance and landing page experience split the rest.

The Most Important Thing to Know First

The score you see in the account is not the value used in the auction. Google itself spells this out in the official Quality Score help docs: the visible 1-to-10 number is a rolling diagnostic, aggregated from historical data. For each individual auction, Google recalculates quality on the fly, with extra signals like device, location and time of day. That per-query quality then combines with your bid to form your ad rank.

Two practical consequences:

1. You can have two keywords with identical QS=5 that perform completely differently in the auction.
2. A QS improvement from 4 to 6 tells you the diagnostic looks better. It doesn’t necessarily move your actual auction performance by the same amount.

If you don’t get that, you’re optimizing a dashboard number and wondering why CPC and conversion rate barely move.

Smart Bidding Changed the Mechanics

Until roughly 2020 the rule was simple: high QS = cheaper CPC. The old rule of thumb was everywhere (“QS 10 saves up to 50% vs QS 5”). Smart Bidding broke that direct coupling.

Smart Bidding does not use the visible Quality Score as input. It uses the underlying signals (ad relevance, expected CTR, landing page experience), your conversion data, and hundreds of other context variables to decide in real time whether and how much to bid. Optmyzr co-founder Frederick Vallaeys, who worked on the original Quality Score system at Google, sums it up in his current take: in 2026, Quality Score is a diagnostic, not a lever.

What stays:

• The three components still feed the algorithm.
• Poor ad relevance or weak landing pages still bite, just differently. Your ad gets served less often, and CPCs creep up indirectly because the model trusts your setup less.

What changes:

• “QS optimization” as a standalone project doesn’t really work anymore. You optimize the content behind QS, not the number itself.

Outside agencies that sell QS optimization as a packaged service usually skip over this shift. It sounds like a clear lever, but in 2026 it’s a lever that no longer directly moves results in most accounts.

Why “Sort by QS” Is the Wrong Approach

You have 800 active keywords. 47 of them have a QS under 6. Standard advice: sort by QS ascending, start with the 1s and 2s.

The catch: a QS-1 keyword with 50 impressions a month might burn $5 a month. A QS-4 keyword with 5,000 impressions at a $1.50 average CPC burns hundreds of dollars, even if the score at first glance looks less dramatic.

The Impact-Score Formula

The lever I use for every QS optimization is a simple multiplication:

Impact Score = Impressions × (10 − QS) × CPC

Two quick examples (hypothetical, not from real accounts):

Keyword A: QS=4, 5,000 impressions/month, $1.50 CPC → 5,000 × 6 × 1.50 = 45,000 impact points

Keyword B: QS=2, 200 impressions/month, $0.80 CPC → 200 × 8 × 0.80 = 1,280 impact points

Keyword A has the lower score, but 35x more impact. Sort by QS ascending and you start with keyword B, ignoring the real problem.

Sort by impact score, descending, instead. The list looks completely different, and you end up working on the keywords that actually cost money.

How to Build the Impact Score in Practice

The standard Google Ads UI doesn’t let you add a custom multiplication formula as a column. There are four ways to handle it, depending on your setup and preferences.

Tier 1: Sheets export (anyone, no setup) Download the keyword report from Google Ads as CSV, or open it directly in Google Sheets. Add a column: =Impressions * (10 - Quality Score) * Avg. CPC. Sort descending. For an account with a few hundred keywords this is plenty. Upside: no technical hurdle, anyone can do it.

Tier 2: Custom column in the UI (a bit of clicking) Google Ads supports custom columns. In the report editor: “Columns” → “Modify columns” → “Custom columns” → new column with the formula. Once it’s saved, you can sort by it.

Tier 3: Google Ads Script (free, no API access needed) Inside the Google Ads account under “Tools” → “Bulk actions” → “Scripts” you can run JavaScript scripts directly. A 30-line script pulls the keyword report, calculates the impact score and writes the top 50 to a Google Sheet. Runs daily or weekly. Worth it once you’ve got multiple accounts or several thousand keywords.

Tier 4: Google Ads API (for developers or agencies) If you’re already using the API, pull the keyword report via GAQL and sort programmatically. Overkill for most solo advertisers, useful for agency tooling.

What you actually use depends on account size. With a few hundred keywords the sheets export is enough. Multiple thousand keywords across multiple accounts is where the script pays off.

What to Actually Do per Component

Ad Relevance Below Average

The keyword doesn’t match the ad copy. Most common cause: too few RSAs per ad group, or the keyword’s in an ad group that’s too broad thematically. Fix:

• Strengthen ad copy by working the keyword (or a close variant) into Headline 1 or 2.
• If the keyword sits in an ad group with ten other themes, split the ad group.
• Asset pinning is the emergency fix, not the first move. Smart Bidding prefers flexibility.

Expected CTR Below Average

Your ad looks less attractive than your competitors’. Fix:

• Pull up competitor ads via the Google Ads Transparency Center.
• Headlines with a clear benefit, not generic filler.
• Use sitelinks, callouts, structured snippets. They lift visibility, and CTR with it.
• Check match type: if broad match is showing your keyword on searches that don’t fit, CTR drops.

Landing Page Experience Below Average

The destination page doesn’t match the promise, or it loads too slowly. Fix:

• Keyword in the H1 and meta description of the landing page.
• Check mobile page speed with PageSpeed Insights or Lighthouse. Scores under 50 are a problem in most verticals.
• If the keyword points to a generic service page instead of a specific one, build a dedicated landing page.
• Landing page experience refreshes slowly. I usually see the score improve 2 to 4 weeks after the change.

When QS Optimization Isn’t (Anymore) Worth It

The honest skip-list I use when prioritizing:

1. Brand campaigns: Searches on your own brand name almost always have high QS. When they don’t, it’s usually a domain or landing-page issue, not a QS issue.
2. Very small accounts: Under 1,000 impressions a month, there’s not enough data to take QS seriously. Better spend the time on match types, negatives and clean tracking.
3. Smart Bidding with enough conversion data: If the account is running profitably, the algorithm has clear signals and CPCs are stable, direct QS improvements get small. Put that hour into negatives or landing-page CRO.
4. Accounts without conversion tracking: As long as you don’t measure what works reliably, QS optimization is just optimizing on gut feel. See why your Google Ads conversion tracking is probably wrong and the common SMB mistakes wasting budget.

Understanding Quality Score is more valuable long-term than maximizing it. If you can read the diagnostic, you can run the account yourself, without leaning on tools or external reports.

FAQ

What’s the difference between Keyword Quality Score and Ad Strength?

Keyword Quality Score rates each search keyword 1 to 10, based on ad relevance, expected CTR and landing page experience. Ad Strength is a separate rating for Responsive Search Ads (RSAs). It assesses how many headlines and descriptions you’ve used and how varied the content is. The two are related but not the same.

Do conversions affect Quality Score?

No. Google states this explicitly in the official ad quality docs: conversions are not a direct input into Quality Score. Your tracking setup is still critical, because without conversions Smart Bidding doesn’t work, and the whole QS-component lever loses its grip.

What does the “rarely shown: low Quality Score” warning mean?

This warning shows up on very low scores (typically 1 or 2). Google is telling you the keyword almost never has a chance in the auction. Options: rework the ad copy, check match type, or remove the keyword entirely if it doesn’t match search intent.

How fast do QS improvements show up?

Ad relevance and expected CTR refresh relatively quickly, often within a few days of an ad change. Landing page experience takes longer. In my experience: 2 to 4 weeks before changes show up in the score.

In Short

If you feel like your account has been “working on QS” for months without any visible CPC or conversion impact, an audit can tell you why: wrong prioritization, Smart-Bidding realities, or a tracking problem sitting underneath the score. On my projects this is part of the 90-day setup, with clear impact-based prioritization, written documentation, and no lock-in. → Book a free intro call