Search visibility: how to measure it, and why rankings aren’t the whole picture

What search visibility means

Visibility answers the question of how noticeable a site is in search for the queries that matter to it. You’ll also see other names: «visibility index,» «visibility score.» These are all the same kind of calculated metric, just from different tools.

Visibility is a metric from SEO tools, not from search engines

Google and Yandex give a site owner factual data on how the site has appeared in search results.

Search Console’s Performance report has four main metrics:

• Impressions — how many times the site appeared in search results. An impression is usually counted when a results page containing it is opened, even if the user didn’t scroll down to it.
• Clicks — how many times users went to the site from Google’s search results.
• CTR (click-through rate) — the share of impressions that ended in a click: the number of clicks divided by the number of impressions.
• Average position — the average ranking of the site’s highest-ranking result in the SERP.

Report data is aggregated by property. If two results from the same site appear for one query, that counts as a single impression.

Webmaster gathers the same kind of data in its query monitoring report. Per Yandex’s help, an impression is a link to the site appearing in search results for a query, and a click is a visit to the site from the results page. CTR there is the ratio of clicks on a snippet to its number of impressions, as a percentage. A snippet is the description of a result in the SERP: the title, text, and address.

Position in Webmaster is the place of the site’s link in the SERP. If several pages of the site are relevant to a query, their impressions and clicks are summed, and the position is averaged. There’s also a «Demand» metric: it shows how often Yandex users make that query.

There’s no metric named «visibility» in these reports. Yandex uses the word in a general sense: per its help, query monitoring helps you keep track of a site’s visibility in search. Search Console’s help on the Performance report doesn’t use the word «visibility» at all.

The conclusion: visibility as a percentage or a score is a derivative of rankings and demand, calculated by a tool against its own set of queries. If you want to know your site’s visibility in Yandex, in Webmaster itself that’s impressions, positions and demand for your site’s queries.

Visibility of a site, a section, and a group of queries

One number for the whole site explains little. A site can grow in search for one section while losing ground in another, and the overall visibility will barely change.

It’s more useful to calculate visibility by parts:

• by group of queries — for example, by all the queries about one service. What’s called «keyword visibility» is exactly visibility by a group of queries;
• by section of the site — a catalog, a blog, a services section;
• by page type — product pages, categories, articles.

The number depends on which queries are included in it. So you first need a set that reflects your business’s demand. The article on the keyword map covers how to build the set of queries you’ll track a site by.

How visibility is calculated: the general principle

What the metric is made up of

Tools’ methods differ, but they usually account for the same components:

1. The set of queries the site is checked against. Set by the tool or by the user.
2. The search volume of each query — how many times it’s searched over a period. This is the demand weight: a ranking for a frequent query counts for more than one for a rare query.
3. The site’s position for each query at the moment of checking.
4. The position weight: the higher the site in the SERP, the bigger that query’s contribution to the total.
5. Check conditions: region, device type, date.

One common approach is to sum the contributions of all the queries and compare the total against the maximum the site would get for first place on every query. The result is then a percentage. Other tools express it in arbitrary points.

In essence, it’s a numerical estimate of how many impressions a site could get for a particular set of queries. Visibility doesn’t measure actual impressions: it’s a model built on rankings and an estimate of demand.

Why different tools give different numbers

What differs in the method	Why the numbers diverge
Set of queries	Each tool has its own database, each project its own list. Queries not in the set aren’t counted
Search-volume source and how recently it was updated	Different demand estimates give the same queries different weight
Position weights	One tool emphasizes top rankings more heavily, another spreads the weight more evenly
Depth of the SERP checked	Positions beyond the checked depth aren’t counted
Region and device	The SERP can differ by region and by device
SERP blocks	A tool may or may not count maps, video, products and other blocks
Scale	Percentages and points from different tools don’t convert into each other

Even the search engines themselves calculate position and impressions differently. Search Console takes the position of the site’s highest-ranking result, and counts two results from the same site for one query as a single impression. Webmaster sums the impressions of several relevant pages and averages the position.

So compare the trend within one source and one method, rather than comparing numbers between different tools.

Sometimes the median position is looked at instead of the average. The median is the middle value if you order the positions across all queries. It’s less affected by a few queries with very low positions.

Visibility, rankings and traffic: what’s the difference

Metric	What it shows	Where the data comes from	What it doesn’t show
Visibility	An estimate of how noticeable the site is for a chosen set of queries	An SEO tool’s calculation	Actual impressions and click-throughs; queries outside the set
Ranking for a query	The site’s place in the SERP for one query at the moment of checking	Manual check or a rank-tracking service	What users see in other regions, on other devices, with a different search history
Impressions and average position	How many times the site actually appeared in the SERP across all queries, and at what average position	Search Console, Webmaster	How many people visited the site
Clicks and CTR	How many times users came from the SERP, and what share of impressions ended in a click	Search Console, Webmaster	What users did on the site
Organic traffic	Visits to the site from regular, non-paid search results	Web analytics systems	Impressions that didn’t lead to a visit

Organic traffic is visits from regular, non-paid search results. Google Analytics’ help describes the «Organic Search» channel as visits to the site via a non-paid link in regular search results. Paid search is visits via ads on search sites.

Visibility is about the chance to catch a user’s eye; traffic is about actual visits. Between them sit clicks: a site can appear often and still get few click-throughs. Visibility by itself isn’t a result: there are still clicks and on-site actions before you get to inquiries.

Why a site ends up at one ranking or another is a separate topic. The article on SEO covers how a search engine finds and ranks pages.

How to estimate visibility from your own data

Impressions, clicks and average position in Google Search Console

The foundation is the Performance report for search results. Its data can be broken down by query, page, country, device, search appearance, and date. The report takes a date range and a search type. The date filter has a comparison mode, for example comparing the current week with the previous one.

The site’s average position is averaged across all queries. This matters when reading the report: if the site starts appearing for new queries at distant positions, the average position can worsen even though visibility has grown.

How to read the data:

• Impressions grow, average position barely changes — the site has started appearing for more queries, or demand for the existing queries has grown. Check the queries tab to see which ones are new.
• Average position improves, but impressions don’t grow — the site has risen for low-demand queries, stopped appearing for queries at distant positions, or gained on some queries while losing ground on others.
• Impressions drop across all pages at once — compare with the same period last year, if you have the data: it may be a seasonal drop in demand.

Keep a limitation in mind: per Search Console’s help, anonymized queries aren’t counted when filtering by query. So the sum for a filtered group of queries can be smaller than the site’s overall total.

Queries and impressions in Yandex Webmaster

In Webmaster you can see which queries the site already appears for in Yandex’s search results. A separate article covers how to connect a site to Webmaster and find the query report; here is how to read that data as visibility.

• Query groups. The report can be filtered by a group of queries you set up in advance. It’s convenient to set up groups for sections or services and look at each one’s visibility separately.
• Region and device. You can set a region and a device type for the analysis. Compare periods under the same settings.
• Impressions and demand. Yandex’s help recommends comparing the trend in impressions and demand to see whether a change in visibility is caused by a seasonal change in demand. If both impressions and demand dropped, it’s most likely the season. If demand is unchanged but impressions dropped, the cause is more likely to be in the rankings.
• More impressions than demand. This happens when the site appears more than once on the first page of results.

Your own visibility estimate by query group

The most reliable approach is to build your own estimate from the search engines’ own data. The process:

1. Choose the queries that matter to the business. Don’t include brand or random queries the site shows for without any benefit. The article on keyword research covers how to select queries and estimate their search volume.
2. Break the queries into groups by section or page type. Account for the user’s task: the article on intent covers why queries with different tasks matter differently for a site.
3. Fix the region and device type. If conditions change, start a new comparison series.
4. For each group, take the impressions and average position from Search Console and Webmaster. Don’t add the two systems’ data together or compare it directly: they calculate impressions and position differently.
5. Compare the same periods, accounting for seasonality: week over week, or a month against the same month last year.
6. When something changes, look for the source: which group and which pages drove the growth or the drop.

This way, visibility stops being a single number from someone else’s tool. It turns into a set of series that show what changed and where. When there are many query groups and sections, tracking like this becomes part of systematic work on a site’s visibility.

Rank-tracking and visibility services

Rank-tracking services pull their own SERP snapshots for a list of queries and calculate visibility. They’re useful when you need rankings for specific queries on each date, or an estimate for someone else’s site. Before trusting a number, find out:

• which set of queries is used — the tool’s own database or your list;
• which region and device the SERP is pulled from;
• how often rankings and search volume are updated;
• how SERP blocks are counted: maps, video, products;
• whether you can export positions for each query, not just the final number.

If the method’s description doesn’t answer these, it’s hard to tell exactly what that number measures.

Why rankings aren’t the whole picture

A ranking depends on the user, the region, and the device

Google, by its own description, uses location, search history and settings to determine what’s most relevant to a particular user. Its own example: a search for «football» is more likely to show American football in Chicago, and soccer in London. Search Console’s help warns that, when you search it yourself, the ranking can differ from the report’s average due to search history or location.

Yandex’s material for business on checking rankings states that results can be affected by the user’s region, device type, time of the check, and SERP personalization.

The practical takeaway: a ranking in your own browser isn’t a visibility metric. It’s one SERP for one user in one place at one moment.

Other blocks can appear above organic results

Organic results are the regular, unpaid links in the SERP. Above and between them you’ll often find ads, quick answers, maps, products, video. Which blocks appear depends on the query.

In Search Console, every SERP element takes up one position, no matter how many links it contains. So at the same position, a site can end up higher or lower on the screen, depending on what’s above it.

Impressions without clicks

Sometimes visibility is high but clicks are few. That usually means the snippet isn’t convincing, or the page is appearing for queries it doesn’t answer well. That’s a reason to check how a snippet affects the decision to click, and to match the page’s content against the query’s task.

Visibility in generative answers

AI-powered answers in search change what a user sees before the list of links. A separate article covers how a site is represented in generative search answers. Search Console has a separate report for this: it gathers impressions in generative-AI-based features — AI Overviews and AI Mode. Webmaster has a «Site visibility in Alice AI» report; Yandex collects statistics there only for queries where the site is already ranking high enough in the SERP. This is data about AI answers, not a calculated visibility score for regular search results.

Finally, any visibility has one precondition — indexing: pages that aren’t in the index don’t appear for any query.

Common mistakes

1. Comparing visibility from different tools. The sets of queries, position weights and scales differ, so the numbers aren’t comparable.
2. Looking at a single overall number. Growth in one query group can mask a drop in another.
3. Judging visibility by a ranking in your own browser. The SERP depends on location, search history and device.
4. Including brand and non-target queries in the set. The number grows, but there’s no benefit for the business.
5. Changing the query set and comparing against the previous period. The change will show a new set, not a new position for the site.
6. Treating a rise in visibility as a result in itself. Without clicks and on-site actions, it brings nothing.
7. Ignoring seasonality. A drop in visibility during a low season can be a drop in demand, not in rankings.

Frequently asked questions

What is a visibility index, and what’s it measured in?

That’s what SEO tools call their own calculated visibility metric. Each one has its own scale: a percentage of the maximum possible value, or arbitrary points. You can’t convert one scale into another: they’re based on different sets of queries and position weights.

Can you compare a site’s visibility from two different tools?

No, not if the set of queries and the method differ. Compare the trend within one source under the same settings. What exactly differs is covered in the section on why tools give different numbers.

Why did visibility drop if nothing changed on the site?

Possible causes: a seasonal drop in demand; a change in the SERP — new blocks or competitors appeared; the tool updated its query set or search-volume data; some pages dropped out of the index. Start by comparing impressions and demand in Webmaster and checking indexing. If the cause isn’t obvious, it may be worth having an SEO audit look into the reasons behind a visibility drop.

Can you find out competitors’ visibility?

Tools estimate the visibility of any domain against their own query database, with the same methodological limits. Detailed Search Console and Webmaster data is only available for sites with verified ownership, so you can’t look at someone else’s site in them.

How often should you check visibility?

That depends on how quickly the site changes and how seasonal your demand is. Day-to-day fluctuations aren’t a reason to draw conclusions: even the time of the check affects the results. Look at the trend over comparable periods.