Google Ads Quality Score Calculator
Estimates quality score from its three components and shows what each score pays for the same position, using the Ad Rank relationship.
Estimated quality score 6 out of 10
from the components Google does not publish the formula, so this is the usual reconstruction
The three components
expected ctr above average
measures whether people click your ad for this keyword, relative to others in the same position
ad relevance average
measures whether the ad text matches the intent of the query, not whether it contains the keyword
landing page below average
measures whether the page delivers what the ad promised, loads quickly and works on a phone
What it does to the auction
your maximum bid $3.50
Ad Rank, roughly 21.00
which is bid times quality, and the real thing includes context and extensions
competitor Ad Rank 14.00
you pay, roughly $2.34
saving against your bid $1.16
position above them
The same auction at every score
score 10 Ad Rank 35.0, pay $1.41
score 9 Ad Rank 31.5, pay $1.57
score 8 Ad Rank 28.0, pay $1.76
score 7 Ad Rank 24.5, pay $2.01
score 6 Ad Rank 21.0, pay $2.34 ← yours
score 5 Ad Rank 17.5, pay $2.81
score 4 Ad Rank 14.0, pay $3.50
score 3 Ad Rank 10.5, pay $3.50
score 2 Ad Rank 7.0, pay $3.50
score 1 Ad Rank 3.5, pay $3.50
Over 2,400 clicks
at your score $5,624.00
at a 6 $5,624.00
at a 10 $3,384.00
the spread $2,240.00
Quality score is a diagnostic, not a lever. It is reported one to ten
from three components, and the thing that actually decides whether you
show and what you pay is Ad Rank, which includes the bid, ad quality at
auction time, the search context, the expected effect of extensions and
format thresholds. The estimate above is the usual reconstruction rather
than a published formula.
Expected click-through rate is the component that moves most, and it is
relative: whether people click your ad for this query compared with
others in the same position. That is why a new keyword starts at an
assumed average and why pausing a poor ad improves the account rather
than hiding a problem.
Ad relevance is about intent, not keyword stuffing. An ad containing the
keyword three times and answering a different question scores worse than
one that does not repeat it and matches what the searcher wanted.
Landing page experience is the component most often ignored and the
cheapest to fix. Load time, mobile usability, and whether the page
delivers what the ad promised. Sending paid traffic to a homepage is the
standard version of getting this wrong.
At a score of 6 you pay about $2.34 where a score of 10 would pay $1.41
for the same position. That relationship, paying the rank you need to
beat divided by your own quality, is the mechanism worth understanding
even though the real auction has more inputs.
The score is averaged and lagged. It is calculated per query and rolled
up to the keyword over a window, so a change today shows up over days,
and a keyword with little traffic keeps a score based on almost nothing.
A score of 10 is not the goal. On a broad keyword it may be unreachable,
and the spend is better aimed at the keywords where a point of
improvement moves real money. Chasing the number itself is how accounts
end up with excellent scores on keywords nobody searches.
Quality score does not exist on every platform, and where it does the
definition differs. Microsoft Advertising has its own version with
different components, and the social platforms have relevance scores
that behave differently again.
Output is valid and updates as you type.
Fix the highlighted fields to update the output.
Quality score is a diagnostic, not a lever. It is a one-to-ten summary of three components, reported after the fact, and the thing that actually decides whether your ad shows and what it costs is Ad Rank: the bid, the ad quality at auction time, the search context, the expected effect of extensions, and thresholds for the format.
What is real is the consequence. A higher quality score pays less for the same position, because in the classic relationship you pay the rank you need to beat divided by your own quality. On the example below, the same 2,400 clicks cost $5,624 at a score of 6 and $3,384 at a 10.
Google does not publish the formula that turns the three components into the number, so the score here is the usual reconstruction rather than a claim.
How to use
- Set each component to what the platform reports: below average, average, above average.
- Put in your maximum cost per click.
- Add the Ad Rank you need to beat to see what each score would pay.
Example
Estimated quality score 6 out of 10
from the components Google does not publish the formula, so this is the usual reconstruction
The three components
expected ctr above average
measures whether people click your ad for this keyword, relative to others in the same position
ad relevance average
measures whether the ad text matches the intent of the query, not whether it contains the keyword
landing page below average
measures whether the page delivers what the ad promised, loads quickly and works on a phone
What it does to the auction
your maximum bid $3.50
Ad Rank, roughly 21.00
competitor Ad Rank 14.00
you pay, roughly $2.34
The same auction at every score
score 10 Ad Rank 35.0, pay $1.41
score 6 Ad Rank 21.0, pay $2.34 ← yours
score 4 Ad Rank 14.0, pay $3.50
Over 2,400 clicks
at your score $5,624.00
at a 10 $3,384.00
the spread $2,240.00Pitfalls
The score is a summary, not an input. Improving it is really improving the three components, and the components are what the platform tells you about. A campaign optimised for the number rather than the components is optimising a report.
Expected click-through rate is relative. It compares your ad against others in the same position for the same query, which is why pausing a poor ad improves the account and why a brand-new keyword starts on an assumed average.
Ad relevance is about intent, not keyword density. An ad that repeats the keyword three times and answers a different question scores worse than one that never repeats it and matches what was wanted.
Landing page experience is the cheapest to fix and the most ignored. Load time, mobile usability, and whether the page delivers the promise. Sending paid traffic to a homepage is the standard way to score badly here.
The score is averaged and lagged. It is calculated per query, rolled up to the keyword over a window, so today’s change appears over days, and a low-traffic keyword carries a score based on almost nothing.
A 10 is not the goal. On a broad keyword it may be unreachable, and effort is better spent where a point of improvement moves real money. Perfect scores on keywords nobody searches is a common and useless outcome.
Other platforms are not the same. Microsoft Advertising has its own version with different components, and the social platforms have relevance scores that behave differently again.
Compatibility
Arithmetic in the browser: nothing is uploaded and nothing is stored.
The score is estimated from the three components with expected click-through rate weighted highest, which every public account of the mechanism agrees on, and the output says plainly that the formula is not published. All-average gives a 6, all-above gives a 10, all-below gives a 1.
The price at each score uses the classic relationship: the Ad Rank you need to beat divided by your own quality score, plus a cent, capped at your maximum bid. The cap matters and is checked in the tests: you never pay more than you bid, however low the score.
The real auction includes the search context, device, location, time of day and the expected effect of extensions, none of which is here. Treat the figures as the shape of the mechanism rather than a prediction of your invoice.