Google Visibility Score: How to Build Yours From Search Console

Google does not publish a visibility score. Search Console has no metric by that name, so every Google visibility score you see in a rank tracker is that vendor's estimate, built from its own keyword list, rank checks and click-through curve. You can build a more honest search engine visibility score from your own Search Console data with one export, your own click curve and a single SUMPRODUCT formula.

What follows: how vendor scores differ from a Search Console score, the formula, a worked example, what a good score looks like, and the four places the number will mislead you if you are not watching for them.

Does Google have a visibility score?

No. Google Search Console reports clicks, impressions, click-through rate and average position in its Performance report, but it does not combine them into a single visibility number. Any score labelled "Google visibility" comes from a third-party tool that models Google's results.

Almost every one of those tools uses the same calculation. Take a list of tracked keywords, find where you rank for each, convert each position into an expected click-through rate, weight by search volume, and add it up. Keyword.com describes its score as built from "your keyword rankings, the click-through rates associated with each position, and the search volumes of your tracked keywords." Moz's help team has described its Search Visibility score in similar terms: an expected CTR is applied to each keyword's position, the CTRs are summed, and the total is divided by the number of tracked keywords. Same idea, different weighting.

Which visibility score should you use?

Use a Search Console score when the question is "are we capturing more of the clicks we could get in Google", and a vendor score when you need to compare yourself with competitors on the same keyword set. They answer different questions, and neither covers AI answers.

Score	Built from	Best for	Limits
Rank-tracker visibility score (Keyword.com, Moz and similar)	The vendor's rank checks, an industry-average CTR curve, third-party search volume	Comparing your site with competitors on one shared keyword list	Moves when the vendor changes its list, location or curve; not your real click behavior
Search Console visibility score (the method below)	Your own positions and your own CTR per position, from Google's data	Tracking your own share of available clicks month over month	No competitor view; only sees queries where you already earn impressions
Raw Search Console averages	Average position and total impressions in the Performance report	A quick health check	Averages hide which queries moved and how much each one is worth
AI visibility score	Whether answer engines mention or cite you, or how ready your pages are to be cited	ChatGPT, Perplexity, Gemini and AI Overviews	Says nothing about classic Google rankings

For a wider look at the tool categories behind these numbers, see our comparison of the four kinds of SEO visibility tool.

Why does a vendor's visibility score move when your traffic does not?

Because the recipe is shared but the inputs are not: whose keyword list, whose CTR curve, which city and device, how often ranks were checked. Change any one of those and the score moves while your traffic stays exactly where it was. We broke those disagreements down in what a search engine visibility tool actually measures. The short version: a vendor score describes the vendor's model about as much as it describes your site.

What Search Console data do you need?

You need three things from the Performance report: position per query, clicks and impressions. Google's documentation on impressions, position and clicks defines position as "the topmost position occupied by a link to your property" in the results, averaged across impressions. From that one report you get:

• Position per query, measured from real impressions across real locations and devices rather than a crawler checking from one data center.
• Your own CTR at each position, from clicks divided by impressions. This replaces the industry-average curve every vendor plugs in with the curve your titles, brand and SERP features actually produce.
• Demand signals, through impressions, for the queries where you already show up.

What Search Console cannot give you is demand for queries where you never appear, because a query is only logged once you earn an impression. Fill that gap with monthly search volume from a keyword tool; the free CrawlTide Keyword Generator returns monthly search volume, CPC and difficulty for related keyword ideas from a single seed term.

How do you calculate a search engine visibility score?

Divide the clicks your current positions earn by the clicks you would earn ranking first for every tracked query, with both sides weighted by demand and priced with your own CTR curve. Five steps:

1. Freeze a keyword set. Pick 30 to 100 non-branded queries that map to pages you want to win. Write the list down and do not edit it between measurements. A list that changes month to month is the most common reason a visibility score jumps while nothing happened in search.
2. Export three months of query data. In Performance, set the search type to Web, choose a three-month range, and export the Queries table with clicks, impressions and position.
3. Build your CTR curve. Using all non-branded queries in the export, not only your tracked ones, group rows into position bands: 1, 2, 3, 4 to 5, 6 to 10, 11 to 20. For each band, divide total clicks by total impressions. Leave branded queries out, because people searching your name click you almost regardless of where you sit.
4. Attach a demand figure to each tracked query. Use monthly search volume from a keyword tool where you have it. Where you do not, fall back to Search Console impressions and mark the row, since impressions understate demand for queries where you rank low.
5. Score it. Visibility = the sum of (demand × your CTR at the query's current position), divided by the sum of (demand × your CTR at position 1). A query where you do not rank contributes zero to the top line and its full value to the bottom line.

In a spreadsheet with demand in column B, your CTR at the current position in column C and your CTR at position 1 in column D, the whole score is one cell:

=SUMPRODUCT(B2:B31,C2:C31)/SUMPRODUCT(B2:B31,D2:D31)

The result is a percentage: the share of available clicks you capture today, compared with ranking first for every query on the list. Because the curve comes from your own site, the score reflects your real click behavior instead of an industry average.

A worked example with three keywords

The numbers below are hypothetical, picked to make the arithmetic easy to follow. Imagine a small SaaS selling software to plumbing businesses, with a CTR curve where position 1 earns 30 percent, position 2 earns 15 percent and position 8 earns 3 percent.

Query	Monthly demand	Current position	CTR at current position	Clicks captured	Clicks available at position 1
crm for plumbers	1,000	2	15%	150	300
plumbing invoice template	400	8	3%	12	120
plumber scheduling app	600	not ranking	0%	0	180

Captured clicks total 162 and available clicks total 600, so this site's visibility score is 162 ÷ 600, or 27 percent.

Now look at what moves it. Getting "plumber scheduling app" onto page one at position 8 adds 18 captured clicks and lifts the score to 30 percent. Moving "crm for plumbers" from position 2 to position 1 adds 150 and lifts it to 52 percent. That comparison is what a visibility score is genuinely good for: it tells you which single ranking change is worth the most, which a list of positions never makes obvious.

What is a good Google visibility score?

A good score is one that rises on the same frozen keyword list; there is no universal benchmark, because the number depends entirely on which queries you chose. A list of hard head terms will produce a low score for a strong site, and a list of easy long-tail terms will produce a high score for a weak one. That is also why scores from two different tools, or two different lists, cannot be compared.

Read the number three ways instead:

• Against itself. A move from 27 to 30 percent on an unchanged list and an unchanged curve is real ranking progress.
• Against the ceiling. 100 percent means ranking first for every query on the list, which almost no site does. The gap between your score and 100 shows how much click demand is still on the table.
• Against the biggest single move. As in the example, the most useful output is often which one query would add the most points, not the headline number.

Where will a Google visibility score mislead you?

A score built this way is more honest than a vendor's, but it inherits the quirks of how Search Console counts. Four matter most:

• AI Overviews collapse into one position. Google's documentation says an AI Overview "occupies a single position in search results," and every link inside it shares that position. A citation in an Overview can log a strong position with a weak click rate, which pulls down the top bands of your curve. If many tracked queries trigger Overviews, compute the curve with and without those queries and compare.
• Average position hides spread. Say a query sits at position 3 on desktop and 14 on mobile. The blended average lands somewhere in between and matches neither result. Segment by device if your audience splits heavily one way.
• Search Labs experiments are invisible. Google states that Search Console does not include data from experiments in Search Labs, so any exposure you get there never reaches this score.
• Sparse bands produce noisy CTRs. A band with a few dozen impressions can show a click rate that swings wildly month to month. Merge thin bands into their neighbors instead of trusting a percentage built on almost no data.

How often should you recompute it?

Monthly suits most sites, because rankings move daily and a monthly figure smooths that noise into a trend you can act on. Keep the CTR curve fixed for a full quarter so that changes in the score come from ranking changes rather than curve changes. When you refresh the curve, note the date in your report, the same way you would mark a methodology change in any other metric.

Weekly makes sense only during a migration or a redesign, when you are watching for a specific drop. Outside those windows, weekly numbers mostly measure volatility.

How do you check search engine visibility for free?

Use the two free webmaster consoles plus the spreadsheet above. Google Search Console gives you the clicks, impressions and positions this method needs. Bing Webmaster Tools gives you the same kind of search performance data for Bing and, since a February 2026 public preview, an AI Performance report showing how often your pages are cited as sources in Microsoft Copilot and Bing's AI-generated answers. Neither console gives you a competitor view; that is the one thing paid rank trackers add.

What does a Google visibility score not see?

It does not see AI answers. Everything above measures Google's results as Search Console reports them, so it says nothing about whether ChatGPT, Perplexity or Gemini cite your pages when someone asks them a question. That is a separate measurement with its own signals, explained in what AI visibility is and how to check your site. If you want the AI-side number broken down the same way this post breaks down the Google one, our guide to how an AI visibility score is calculated explains why two tools give two different numbers.

Treat the two as companion numbers. Your Google visibility score tells you how much of the available click demand you are capturing in Google. An AI visibility score tells you whether answer engines mention you, or whether your pages are easy for them to find and quote. A site can be strong on one and invisible on the other.

Freeze the list, build the curve, then check the AI side

Start with one export and a frozen list of 30 queries. Build the curve, run the formula, and record the number with the date and the list it came from. Next month, the change will mean something, because nothing but your rankings will have moved.

For the half of visibility Search Console cannot see, run your most important page through the free AI Visibility Check. It scores a single URL from 0 to 100 on readiness signals such as entity clarity, structured data and crawlability, and lists the fixes in order of impact. It does not send prompts to any AI engine, so treat it as a readiness check, not a citation count.

If you would rather not rebuild the export by hand each month, CrawlTide's keyword tracking imports queries from a connected Google Search Console property, so your frozen list and its positions stay in one place.