Google Search Console terms, explained.

Search Console is Google's free tool for seeing how your website appears in search. This page explains its screens, labels and terms simply, and which of them you actually need to worry about.

01. Setting up and access

Google Search Console.

Google Search Console is a free tool from Google that shows how your website appears in Google Search. It reports which searches your pages appeared for, how many people clicked, which pages Google has stored, and any problems it has found.

It only shows figures for a website once someone has proved they own it, so it is never public. It is the source of the search figures in your monthly report, and you can be given a view of it without handing over any control.

For example, a plumber who has never opened Search Console can still be sent a report built from it. If they want to look, they are added as a user on their own property and see exactly what we see.

See also Search performance dataAccess requests

Domain property.

A domain property covers your whole web address in one go: the version with www and without it, http and https, and every subdomain such as shop.yourbusiness.co.uk. It is the most complete way to set Search Console up, and it is the one we ask for when we can.

It can only be verified by adding a short record in the settings where your domain name is managed, usually with your domain registrar or hosting company. That is a one-off job, but it is also why it sometimes needs your web developer or whoever looks after your domain.

For example, if your site has moved from http to https and added a blog on a subdomain, a domain property still shows all of it together. A narrower property would miss the old and the new versions.

See also Account ownership

URL-prefix property.

A URL-prefix property covers only addresses that start with exactly what was entered, including the http or https and any folder. So https://yourbusiness.co.uk/ does not include the http version, a www version or a subdomain.

It is easier to verify than a domain property because there are several methods, such as adding a small tag or file to the website. It is useful when you only have access to the website itself, but it can quietly miss part of your traffic if the prefix is not the version Google actually shows.

For example, if the property was added as https://yourbusiness.co.uk/ and Google mostly shows the www version, the figures can look much smaller than reality. Nothing is broken, the property is simply looking at one of several versions.

See also One website address

Verification.

Verification is how Google confirms that the person adding a property really controls the website. Google offers several methods, such as uploading a file, adding a tag to the website, using Google Analytics or Tag Manager, or adding a record to the domain settings.

Once verified, the method has to stay in place. If someone removes the tag or record during a redesign, the owner can lose access later. That is one of the most common reasons a client's Search Console suddenly shows as unverified.

For example, a web agency rebuilds a client's site and deletes the old verification tag along with other old code. A month later the owner finds Search Console asking them to verify again.

See also Making changes safely

Permission levels.

Search Console gives each person one of a few levels. An owner has full control, including adding and removing other people. A full user can see all the data and take some actions. A restricted user can see most data but has fewer rights and only receives messages that affect them directly.

We normally ask to be added as a full user on your property. That is enough to read every report and, with your say-so, to carry out routine checks, and you stay the owner and can remove us at any time.

For example, the business owner is the verified owner, their office manager is a restricted user who just wants to see the figures, and we are a full user who can run the checks that go into the report.

See also Access requestsAccount ownership

Verified owner and delegated owner.

A verified owner proved ownership directly, using one of the verification methods. A delegated owner was made an owner by an existing owner, without needing to verify anything themselves. Both have the same powers.

The difference matters when ownership is passed on. A delegated owner can be removed from the Users and permissions screen, but a verified owner stays until their verification method is taken away from the website.

For example, if you want to be sure your previous agency can no longer control your property, it is not enough to remove them as a user. If they verified it themselves, their tag or record also has to come off the website.

See also Account ownership

02. The Performance report

Performance report.

The Performance report is the main screen most business owners will recognise. A chart across the top shows four measures over time, which are total clicks, total impressions, average click-through rate and average position, and a table underneath breaks them down.

Newer versions of Search Console label this screen Search results, and it sits under Performance in the left-hand menu. The figures are the same ones that appear in our monthly reports, so this is the screen to open if you want to see where a number came from.

For example, a roofer opens it and sees 1,200 clicks and 40,000 impressions for the last three months. The same figures are in the report we send, just explained and compared with the previous period.

See also Search performance dataClicksImpressions

Search type.

The Performance report can be switched between kinds of Google search. Web is the main one, and there are others for Image, Video and News. Each shows only the activity from that kind of search.

Web is the default and the one we report on. Clicks and impressions from Google's AI features, such as AI Overviews and AI Mode, are counted inside the main performance figures. Since 2026 Search Console also has a separate generative AI performance report that shows impressions, pages, countries and devices for those features, but Google lists impressions only, so it cannot tell you how many clicks came from an AI answer, and sites with too few impressions may not see it.

For example, a photographer may see very few clicks under Web but a good number under Image, because people find their work through image searches rather than ordinary ones.

See also AI Overviews

Queries.

The Queries tab lists the words and phrases people typed into Google before your site appeared, with the clicks, impressions, click-through rate and position for each. It is the only place where the real search wording is visible.

A long list is normal. Read it for patterns, such as whether people are searching for what you actually sell, rather than trying to judge it line by line.

For example, an accountant may find that people reach them by searching for "self assessment help Leeds" far more than the phrase on their main page, "personal tax services". That tells us which words to use.

See also Brand versus non brand searches

Pages.

The Pages tab lists your own web addresses, with the same four figures for each. It shows which pages Google is actually using to bring people to you.

Clicking a page and then switching to the Queries tab shows the searches that page appears for. That is how we tell which page is winning the right searches and which one is appearing for something it was never written for.

For example, a dentist's blog post about wisdom teeth may bring in more clicks than the treatments page. That tells us the blog post is worth strengthening and linking onwards to the service page.

See also Internal links

Countries and devices.

The Countries tab shows where in the world the searches came from, and the Devices tab splits them between mobile, desktop and tablet. Both are useful checks that your visibility matches who you actually serve.

A local business with a lot of impressions from another country is not necessarily doing anything wrong, but those figures will push impressions up without bringing customers.

For example, a Cape Town electrician sees 30 per cent of impressions from the United States. They are not customers, so we filter to South Africa before judging whether visibility is growing.

See also Language and regionMobile usability

Search appearance.

Search appearance shows how your result looked in Google, for example whether it appeared with extra features such as review stars, a video or a product listing. It only lists the types that applied to your pages in that period.

If nothing is listed, it usually just means no special result types were shown. It is not a fault on its own.

For example, a bakery with product markup on its cake pages may see a Product results line here. A bakery without it will see nothing, and that is the cue to look at what extra detail could be added.

See also Structured data

Filters and comparisons.

The Performance report can be narrowed with filters, for example to a single page, a single search phrase, one country or one device. It can also compare two periods side by side, such as this month and last month, or the last 28 days and the same days a year earlier.

Most useful questions in Search Console are answered by combining a filter and a comparison. We use them to find pages whose clicks dropped, and searches whose impressions are rising while clicks are not.

For example, comparing this October with last October for one service page shows whether a dip in clicks is the website's doing or just the normal seasonal pattern for that service.

See also Month on monthSeasonality

Data delay and preliminary data.

Search Console is not live. The latest day or two of data is shown as preliminary, drawn with a dotted line on the chart, and it can change within a few hours as Google finishes counting. Most reports also run a few days behind real time.

Because of this we never quote a single recent day in a report. A sudden dip in the last two days of the chart is nearly always the data catching up rather than a real fall.

For example, an owner sees the line drop sharply on the right-hand edge of the chart and worries. A look at it two days later shows the same days are back at their usual level.

See also Clicks

Anonymised queries.

Google leaves out the wording of some searches to protect people's privacy. Google describes them as searches made by only a few dozen people over two to three months. Their clicks still count in the chart totals unless you filter by query, but they never appear in the list of queries.

That is why the clicks in the Queries table can add up to less than the total in the chart. Nothing is missing from your performance, it is simply not shown by name.

For example, a solicitor sees 900 clicks in the chart but only 760 across the listed queries. The other 140 came from rare, one-off searches that Google does not name.

See also Search performance data

Chart and table totals.

The totals above the chart and the totals you get by adding up the table can legitimately differ. The chart counts everything for the property, while the table can group figures differently, for example by page, and leaves out hidden queries.

So two correct numbers can disagree. When we quote a figure in your report, we say which one it came from and use the same one every month, so the comparison stays fair.

For example, the chart says 2,310 clicks and the Pages table adds up to 2,290. The gap is small and expected, and it does not mean either one is wrong.

See also Search performance data

The 16-month limit.

Search Console only keeps about 16 months of performance data. Anything older is no longer available in Search Console, even if you gain access later.

This is why we export your figures and record a starting position at the beginning of the work. It is also why a client who joins us after years of owning a site cannot see how it did three years ago.

For example, a business wants to prove its search traffic was higher in 2023 before a redesign. If nobody saved the figures at the time, Search Console can no longer show them.

See also Search performance dataStarting baseline

Branded queries filter.

Search Console now offers a Branded and Non-branded filter on the Search results Performance report. Google announced it in November 2025, and it is available to eligible sites, meaning top-level properties with enough search volume. It uses a Google system, not a list of words, to decide what counts as your brand, including variations, misspellings and your own products or services.

It is a convenient starting point, but Google says some queries may occasionally be misidentified, so we still check the main branded and non-branded searches by hand before putting a split in a report.

For example, a florist called Rossi Florists sees 'rossi florists' counted as branded and 'wedding flowers Bristol' counted as not. If a search that is clearly about the florist is listed as non-branded, we note it in the report.

See also Brand versus non brand searches

03. The Page indexing report

Page indexing report.

The Page indexing report lists the web addresses Google knows about and whether each one is stored in its search collection. Each address is either Indexed, which means it can appear in search, or Not indexed, with a reason given.

A long Not indexed list is not automatically a problem. Many reasons are perfectly sensible, such as a duplicate page or a thank-you page that was never meant to be found. What matters is whether the pages you want customers to find are on the Indexed side.

For example, a site with 80 pages might show 55 indexed and 25 not indexed. If the 25 are old tag pages, filtered shop pages and redirects, that is healthy. If they include your main service pages, that is the issue to fix.

See also IndexingCrawling

Crawled - currently not indexed.

Google visited the page, read it, and chose not to store it for now. It may well be stored later, and asking Google again does not speed it up.

Google does not give a reason. Pages that are thin, very similar to another page or not obviously useful compared with what is already stored are the usual suspects, so improving the page and linking to it from stronger pages on your site is a sensible response.

For example, a tiling company has ten near-identical area pages that differ only in the town name. Google may store two and leave the rest as Crawled - currently not indexed, until each is given something genuinely local to say.

See also Thin pagesIndexing

Discovered - currently not indexed.

Google knows the address exists, but has not yet visited it. Google says this typically happens when it expected visiting the page to overload the site, so it rescheduled the visit.

A few of these on a new site is not unusual. If an important page stays here, linking to it from a page that matters is a sensible step, though Google does not promise it will visit sooner.

For example, a new service page is listed in the sitemap but not linked from the menu. A month later it is still waiting, until a link from the home page is added.

See also Discovered but not indexedInternal links

Duplicate without user-selected canonical.

Google found two or more pages with the same, or nearly the same, content and nobody told it which one to prefer. Google picked one itself and left the others out.

This is Google doing its job rather than an error, and it is often fine. It becomes a worry only when Google has picked the version you did not want, in which case naming your preferred version with a canonical tag fixes it.

For example, a shop sells a jumper in four colours and each colour has its own address. Google picks one and lists the other three under this label. That is acceptable unless it picked a colour you stopped selling.

See also Canonical tagsDuplicate content

Duplicate, Google chose different canonical than user.

You named a preferred version of the page, but Google decided another one was a better match and indexed that instead. Your named page was left out.

It generally means two pages are too alike, or the signals disagree, for example when the sitemap lists one address while the page itself names another. The URL Inspection tool shows both choices side by side so the difference is visible.

For example, a page says its real version is the address without a trailing slash, but every link on the site uses the one with. Google follows the links and ignores the tag.

See also Canonical tags

Alternate page with proper canonical tag.

This page correctly points to another page as its main version, and Google has accepted that. It is reported as Not indexed because the main version is the one that is stored.

Nothing is wrong and no action is needed. It is the label that appears when canonical tags are working as intended, so seeing a lot of them is usually a sign of a tidy setup.

For example, a page reached through a tracking link such as ?utm_source=newsletter correctly names the normal address as its main version. The tracking one shows up here, which is exactly right.

See also Canonical tags

URL marked 'noindex'.

The page carries an instruction telling Google not to show it in search, and Google has followed it. Older guides may word this label differently.

It is fine for pages that should stay private, such as thank-you pages. It is a real problem when it appears against a page you want found, often because a setting meant for the redesign was never switched off after launch.

For example, a new website goes live with a setting that asks search engines not to index it, left over from the build. Weeks later the service pages are still listed under this label, and the business wonders why it has no enquiries from Google.

See also NoindexStaging sites

URL blocked by robots.txt.

A rule in your website's robots file stopped Google from visiting the page, so it could not read it. Older guides may word this label differently. A related label, Indexed, though blocked by robots.txt, is covered below.

Blocking is intentional for admin areas and baskets. It is a problem where it has swept up pages that matter, and a single wrong line can hide a whole section of the site. Fixing the rule, rather than the page, is what sorts it.

For example, a developer blocks the whole /shop/ folder while building it and forgets to lift the block at launch. Every product page then shows this label.

See also The robots fileCrawling

Indexed, though blocked by robots.txt.

Google could not visit the page because of a robots rule, but stored the address anyway, usually because other pages link to it. It can show in search with little or no description.

If you want the page kept out of search, blocking it in the robots file is the wrong tool. Google has to be able to visit a page to see a noindex instruction on it, so the block must be removed first.

For example, a private price list is blocked by the robots file but linked from a supplier's site. It appears in Google as a bare address with no description, which is the opposite of what was wanted.

See also The robots fileNoindex

Soft 404.

Google visited the page and found what looked like a page-not-found message, or a page with almost nothing on it, but the website returned a normal success code. Google treats it as an error page and leaves it out.

It is commonly seen on empty category pages, search results pages with no matches and removed products. If the page is really gone, it should return a proper not found response. If it should exist, it needs real content.

For example, a shop keeps the page for an out-of-stock range live, showing only the words "No products found". Google files it as a Soft 404.

See also 404 errorsDiscontinued products

Page with redirect.

The address sends visitors on to a different address, so Google does not index the old one. Whether the destination is indexed depends on what Google thinks of that page.

This is expected wherever you have moved or renamed a page. It only needs a look if the redirect leads somewhere unrelated, or if your sitemap still lists the old addresses.

For example, after moving to shorter web addresses, a business sees its 40 old addresses listed under this label. That confirms the redirects are working.

See also Redirects

Not found (404).

Google asked for the page and your website said it does not exist. If you removed the page and nothing replaces it, that is fine and Google will let it drop.

It deserves attention where the page moved and a redirect was missed, or where other sites still link to it. Those visitors and that value are being lost.

For example, a builder renames its extensions page and forgets the redirect. The old address, which a local magazine links to, now shows as Not found (404).

See also 404 errorsRedirects

Validate fix.

After a problem has been fixed on every affected page, the Validate fix button asks Google to check them again. Google then sends a notification with the result, and it typically takes up to about two weeks, sometimes much longer.

Validating does not make Google store a page. It only confirms that the specific problem is gone. If a page was left out for being thin, fixing the wording and validating still leaves that decision with Google.

For example, a business removes a stray noindex from 14 pages and presses Validate fix. A fortnight later a notification confirms the fix worked, and the pages start returning to the Indexed list.

See also Indexing

04. Sitemaps and URL Inspection

Sitemaps report.

The Sitemaps screen is where a sitemap is submitted to Google and checked. It shows when Google last read each one and how many addresses it found.

The status next to each file tells you whether Google could read it. Success is the one you want. Has errors or Couldn't fetch means Google either read it with problems or could not reach it at all, and that needs a look.

For example, a sitemap showing 120 discovered pages for a site with 40 real ones tells us it is listing old, removed or duplicate pages, which is worth tidying.

See also Sitemaps

URL Inspection.

URL Inspection is the search bar at the top of Search Console. Paste one web address into it and Google reports what it knows about that single page: whether it is stored, when it was last visited, which sitemap listed it and which version Google treats as the main one.

It is the quickest way to answer "why isn't this page showing?". Its results describe the last time Google visited, not the live page right now.

For example, a client asks why their new offers page is not on Google. A quick inspection shows it was found but never visited, so the next step is to link to it, not to rewrite it.

See also Indexing

Live test.

Alongside the stored result, URL Inspection can test the page as it is right now. It shows whether Google can reach it today and what the page looks like to Google, which is useful after a fix.

A successful live test does not mean the page will be stored. It only proves that nothing is blocking Google from reading it today.

For example, after removing an accidental noindex, a live test confirms Google no longer sees it. That is worth knowing before waiting weeks to see if anything changes.

See also Noindex

Request indexing.

Request indexing is a button in URL Inspection that asks Google to visit a page sooner. It is useful for a brand new page or an important page that has just been changed.

It is a request, not an instruction. Google may still choose not to store the page, repeating the request does not help, and each property only gets a limited number a day. For lots of pages, a good sitemap and sound internal links work better.

For example, after publishing a new service page we request indexing once. If it is still not stored after a few weeks, the fix is links and content, not more clicks on the button.

See also CrawlingIndexing

05. Experience and enhancements

Core Web Vitals report.

This report grades your pages as Good, Need improvement or Poor, for mobile and desktop separately. It is built from real visits by Chrome users, over the last 28 days, and pages are grouped with similar ones.

A group takes the grade of its weakest measure, and the report only includes pages with enough visits to measure. A small website may show no data at all, which is normal and not a fault.

For example, a group of product pages is marked Poor on mobile because the largest image takes four seconds to appear. The group will take a few weeks to turn green after the image is fixed, as new visits replace old ones.

See also Core Web VitalsPage speed

Good, Need improvement and Poor.

Each of the three Core Web Vitals has two lines. The main content should appear within 2.5 seconds, the page should respond to a tap within 200 milliseconds, and the layout should not jump by more than 0.1. Past those lines a page is Need improvement, and past the second set of lines it is Poor.

Only Poor and Need improvement groups need work. Good means pass, so there is nothing to do for those, however long the list.

For example, a page loading its main image in 2.2 seconds is Good for loading. At 3.5 seconds it would be Need improvement, and over 4 seconds, Poor.

See also Layout shift

HTTPS report.

The HTTPS report shows how many of your pages are served over a secure connection and how many are not. Pages that are not secure are listed with the reason.

It is not a list of every page on your site. A website should serve every page securely, so any not-HTTPS entry on a page that matters is worth fixing.

For example, a few older blog posts still link to images using the insecure form of the address. The report lists those pages, and the fix is to update the image links.

See also HTTPS and SSL certificates

Rich result and enhancement reports.

Search Console adds extra reports for the extra detail it finds on your pages, such as products, reviews, videos or breadcrumbs. Each lists the pages with valid markup, the pages with warnings and the pages with errors.

Only the types Google finds on your site are listed, and Google changes which ones it supports from time to time. An error means that page cannot qualify for the extra display, so we fix errors first and treat warnings as optional.

For example, a bakery's Product report shows 12 valid pages and 3 with errors. The three are missing a price, so they cannot show it in the search result until it is added.

See also Structured dataSchema errors

Retired reports.

Some reports you may see mentioned in older guides no longer exist. Google retired the Mobile Usability report, the Mobile-Friendly Test and its API from 1 December 2023. In the same announcement it said the Page Experience report would become a page linking to its page experience guidance, alongside the Core Web Vitals and HTTPS reports.

Mobile still matters, because Google judges the mobile version of your website. Google points to tools such as Lighthouse in Chrome for checking it, and we use those and our own testing, alongside the Core Web Vitals report.

For example, if a guide tells you to check the Mobile Usability report and you cannot find it, nothing is broken with your account. It has simply been removed.

See also Mobile usabilityCore Web Vitals

06. Security and manual actions

Manual actions report.

The Manual actions report tells you whether a person at Google has decided your site breaks its spam rules, for example for buying links or publishing pages meant only to manipulate results. Most of the time it says No issues detected.

If there is one, the affected pages, or sometimes the whole site, can be held back until it is fixed. It is a rare finding, but it is the first thing we check on a site that has fallen sharply.

For example, a business that bought a package of links years ago finds a manual action listed. Until the links are removed and Google reviews the site, search work will have little effect.

See also Manual actions & security warningsSpam & artificial links

Security issues report.

The Security issues report shows whether Google thinks your site has been hacked or could harm visitors, for example by carrying malware. Affected pages can show a warning in the search results, and browsers can warn people before they open the site.

It is different from a manual action, which is about breaking search rules. A security issue is about visitors' safety. A hacked site often looks normal to its owner, so the report is the place where the problem tends to show up.

For example, a garage's website begins appearing for unrelated medication searches under its own name. The pages were injected by a hacker, and the owner only learned of it from this report.

See also Manual actions & security warnings

07. Links, removals and insights

Links report.

The Links report lists the websites that link to yours, which of your pages receive the most links, and the words used in them. It also shows how your own pages link to each other.

It is a sample, not the full picture. Tables are capped at 1,000 rows, some links may have been removed, and it does not show which links are marked nofollow. We use it as a first look, not as a final count of your links.

For example, the report shows a local newspaper and a trade association among the top linking sites. That is a good sign, and tells us which kind of site is worth approaching for more.

See also BacklinksInternal links

Removals tool.

The Removals tool lets an owner hide a page from Google Search results quickly, for about six months. It is meant for urgent cases, such as something published by mistake.

It does not delete the page from Google permanently, and it does not remove it from the internet. To get rid of a page properly, it must also be taken down, return a not-found response, or be marked noindex, otherwise it comes back when the six months end.

For example, a price list with the wrong prices goes live by mistake. A temporary removal hides it from search the same day while the website is corrected.

See also NoindexRedirects

Search Console Insights.

Insights is a simplified summary report inside Search Console, built to be read without training. Google announced in June 2025 that it was moving into the main Search Console interface, replacing the earlier standalone version. It highlights what is trending up or down, your best-performing pages and the searches bringing people to you, with the key figures taken from the same data as the Performance report.

It is a good starting point for an owner who wants a quick look, but it simplifies. Our monthly report goes further by comparing periods and explaining the reasons behind the changes.

For example, an owner opens Insights and sees that one blog post is trending up. They can then ask us whether it is worth building a service page around the same topic.

See also Monthly reportSearch performance data