How to Read the Search Console Performance Report

To read the Search Console performance report, open it, set the date range and search type, read the four metrics, then group the table by queries and by pages and compare periods. Do it monthly and you will know which searches bring visitors, which pages earn them and where you are seen but not clicked.

This guide is for site owners and marketers who already have a verified property and want to turn its numbers into decisions. Site owners get more from Google Search Console than from most free tools, because it reports Google’s own data about your website, not a third party’s estimate.

It follows Google’s own help pages for every definition, so you can check each one at the source.

If you have not connected your site yet, start with how to set up Search Console, Analytics and AI search tracking. This article picks up where that one stops: your data is flowing, and you need to know what it says.

How to read the Google Search Console performance report in seven steps

Google describes the report as showing important metrics about how your site performs in Google Search results. You can use it to see how search traffic changes, where it comes from and which search queries show your site.

The seven steps below follow the order in which the report is easiest to understand.

1. Open the report and set the date range and search type

In Search Console, open the Performance section, then Search results. The default view shows click and impression data for the past three months.

Two filters are always present: the date range and the search type. Change the date range first, because every number depends on it.

Then check the search type. Web (text-based) covers searches typed into the standard Google search bar. Web (multimodal) covers searches made with images, such as Google Lens. Image, video and news each cover their own tab.

Use the 24-hour view only to watch something that just changed. Google notes that its data points then represent hours and include preliminary data. For everything else, use full days, weeks or months.

2. Read the four metrics: clicks, impression count, click through rate and average position

The report offers four metrics, and you switch each on or off by selecting its tile above the chart. Google defines them as follows:

Position needs care. Google’s page on position explains that 1 is the topmost position, and that the number you see is an average across many searches. Your own search can differ from the average because of your location and search history.

Google’s advice is to focus on trends in impressions and clicks more than on position alone. A high click-through rate, Google says, means your content seems to match what users are searching for.

Some reports show additional metrics, such as query or position, depending on the report type. Start with these four, because they describe your search performance in Google Search results. Other metrics can wait.

3. Read the graph before the table

The chart shows total clicks, total impressions, average CTR and average position for the whole property over the selected time period. The newest data can be preliminary, shown by a dotted line on the graph, because it is still being collected and may change within hours.

Switch the time granularity to weekly or monthly when you want the trend. Google says weekly and monthly views smooth out the swings caused by weekends and holidays, which makes growth or decline easier to see.

Compare the graph with the dates you published a post or changed the website.

Then look for three shapes. A steady climb in impressions means more people are seeing you. A drop that begins on one date points to one cause, such as a site change. A flat line with a falling click count means you are seen as often but chosen less.

4. Group the table by queries to see what people searched

Select the Queries tab. Queries are the search terms that led to your site, and each one is a keyword someone used to find you. Treat the list as keyword research: it shows which phrases already bring search traffic. Click a column heading to sort by clicks, impressions or CTR.

Three limits shape how you read the list:

• Queries shown are exact matches and are not case-sensitive, so similar phrasings appear as separate rows.
• Google omits anonymized queries, which are rare searches, from the table to protect privacy. They remain in the chart totals unless you apply a query filter.
• The table holds at most 1,000 rows, so the long tail is cut off.

Use the branded and non-branded query filter to separate people who know your name from people who found you by searching a topic. The non-branded group is where new audience growth shows up.

Google says the filter starts from March 11, 2025, is unavailable for sub-properties and sites with few impressions, and may misclassify some queries. Your organic keywords are the terms this tab surfaces, and the non-branded rows show new opportunities.

A Botswana safari operator shows why that split matters. Wild For Game Safaris saw organic visits rise from 10 to an estimated 907 a month, all of it non-branded and without a paid click, as its case study records. Only a non-branded view proves that kind of growth came from new readers.

5. Group the table by pages to see what earned the traffic

Select the Pages tab to identify what earned the traffic. It groups data by the final URL linked by a Search result, after any redirects. Choose the Clicks metric and sort to see which pages get the most traffic from Google.

Google assigns most data to a page’s canonical URL, not to a duplicate. If someone clicks a duplicate address, the click counts toward the canonical one. To find a page’s canonical, use the URL Inspection tool.

Now connect the two tabs. Click a specific query to filter the whole report to it, then open the Pages tab to see which URLs Google showed for it.

Do the reverse for a specific page to see every query that surfaced it. Two pages from your site showing for the same query is a signal to check whether they compete.

6. Filter and compare to find what changed

Click Add filter to narrow the data by query, page, country or device, such as mobile, desktop or tablet.

The filter options also include search appearance, which groups data by result type, such as videos, product snippets or rich results built from structured data.

Click any row in a table to filter by that row. The Share button creates a link that gives access only to the current view of the report, not to your property.

You can match items containing a string, or use a regular expression to match several similar queries at once, such as every phrasing of one topic.

Comparison is the most useful feature for a monthly review. Choose Compare in a filter, then pick two date ranges, countries, devices or search types. A Difference column appears beside the table.

Google keeps separate reports for Discover and News, and you cannot combine them in one view in the browser. You can export the data to compare them.

Google recommends weekly or monthly granularity for date comparisons, because a daily comparison can set a Monday against a Saturday. Only one comparison runs at a time, and a new one replaces the old.

Compare the last full month with the previous period first, then the same month a year ago if you have it. Sort the Difference column to find the queries and pages that moved most.

7. Turn what you see into one decision per page

Reading is only useful if it ends in a change. Google’s own use cases give you three starting points.

• Low CTR pages. Select CTR, open the Pages tab and sort ascending. A low rate can mean people see your page but do not think it answers their query. Google suggests updating the title and description, or adjusting the content to match the queries.
• High impressions, low clicks. Sort queries by impressions and look for terms where a post or page appears often but rarely gets chosen. Rewriting its title for more clicks is the cheapest win.
• Position history. Add a URL filter, select Average position only and choose a time frame. Compare it with a similar period before you decide anything.

Google also warns that it is hard to prove a page change caused a result, because news events, shifts in user interest and competitors all move the numbers.

Check that the improvement lines up in time with your edit, compared with similar past periods.

Mark the edit on the chart with a custom annotation: right-click the chart on the date, type up to 120 characters and select Add.

A property holds up to 200 annotations, and notes older than 500 days are deleted. Annotations are shared across the property, so a restricted user has access to view them but not to add or delete.

The generative AI performance report shows where AI features display your pages

Google’s generative AI features are counted in the numbers above, but not separated there. Google says data from AI Overviews and AI Mode stays in the overall performance report.

In June 2026 it announced the new reports as a dedicated view, and as of August 31, 2026 it says they are available to all websites worldwide.

Open the generative AI performance report from the same area of Search Console. It covers impressions for AI Overviews and AI Mode on Google Search. Its dimensions are pages, countries, devices and dates, and the date range and search type filters work as they do in the main report.

It reports impressions only. It does not give clicks, CTR or queries, so do not read it as a traffic report. If you do not see it, Google says your site may not have received enough impressions in these features, or you may have excluded your site from them.

Treat it as a visibility check. It shows which pages Google’s AI features displayed. To measure whether other AI tools name your brand, add the monthly question test described in how to measure search and AI visibility.

The 16-month limit in Google Search Console means you should save your data before it expires

Search Console does not keep your history forever. Google’s Analytics help page states that Search Console keeps data for the last 16 months, and the performance report’s branded filter likewise carries a 16-month history.

That affects year-on-year comparisons. Compare the same month across two years and the earlier month can fall off the edge. Plan around it in three ways:

• Export on a schedule. Use the export button on the report to download both chart and table data, and save a dated copy each month. Values shown as ~ or – in the report appear as zeros in the download.
• Set up the bulk data export. Google lets you schedule a daily export of performance data to BigQuery. It shows all the performance data available to Search Console, apart from anonymized queries. Google says the data accumulates unless you set an expiration time, and that you can see history from before setup only through the API or the reports.
• Record your headline numbers. A one-row summary each month costs two minutes: clicks, impressions, CTR, average position and your top five queries.

Start the export early. Whatever you do not save in month one cannot be rebuilt once month seventeen arrives.

Search Console numbers will not match your Analytics, and that is expected

Site owners often panic when Search Console clicks differ from Analytics sessions. Google lists the reasons on its own help pages. The two tools measure different things:

• Privacy. The report omits some data, such as queries searched very few times.
• Processing. Search Console removes duplicates and other noise, so stats can differ.
• Time lag. Collected data is usually available in two to three days.
• Time zones. Daily data in the report is labeled in Pacific Time, while Analytics uses your own time zone.
• JavaScript. Analytics counts only visitors with JavaScript enabled.

Inside the report, chart and table totals can also disagree. The chart is aggregated by property, the table by page when you group by pages, and the table cuts off at 1,000 rows.

Google’s example: a site with three pages in one result, each clicked once, shows a CTR of 100 percent by property and 33 percent per page.

Use Search Console to see what happened before the click and Analytics to see what happened after. Linking them, as the setup guide shows, lets you follow a query through to a key event.

A monthly reading routine turns the report into a scorecard

Block thirty minutes in the first week of each month, after the previous month’s data is complete. Work through the same list every time so changes are visible.

These numbers are your key performance indicators for organic search.

When the report shows a drop that no content change explains, technical SEO checks come next, and how to maintain your search presence sets out the routine that keeps the basics healthy. For the wider picture of how search works, read the ultimate guide to SEO.

If you would rather have a partner read the report with you every month, the Brand Growth Program is accountability to your Roadmap, measured. Its plans include a monthly report and a quarterly strategy review, with a live dashboard on the higher plans, and entry requires Discovery.

For a weekly note on what is changing in how brands get found, subscribe to Ground Truth, published on Wednesdays.

FAQ