Performance in Google Search Console: Analyse and Take Action in 2026

This page is for SEO managers, marketers and business leaders who want to measure performance in Google Search Console and turn it into priorities: which pages to rework, which queries to target, which drop to tackle first.

It covers the Performance report in detail: its four metrics, its filters, its search types, period comparison, the Search Console Insights report and where AI Overviews fit in. For a broader overview of the tool, read our main guide to Google Search Console.

Tracking Performance in Google Search Console: Analysing and Steering Your SEO Results

What the Performance Report Measures and When to Use It

The Performance report reveals what is happening in Google Search results. It aggregates clicks, impressions, click-through rate (CTR) and average position, broken down by query, page, country, device and search appearance. Use it to:

• identify editorial opportunities;
• analyse traffic and visibility changes;
• prioritise high-impact optimisations;
• track trends and seasonality.

Performance can no longer be reduced to traffic: impressions measure your presence on high-volume queries, even when no click follows.

Where to Find the Report in Google Console and Configure It Correctly

In the Search Console menu, open Performance then Search results. Before analysing, check three points:

• the selected property matches the version of the site you manage;
• the date range is long enough to smooth out daily variations;
• all four metrics are enabled above the chart.

Then segment by country and device. If the property does not exist yet, creating and verifying it (DNS for a Domain property, several methods for a URL-prefix property) is covered in our guide to verifying a property in GSC. Wait a few days of data collection before interpreting the initial data.

Interpreting Metrics Without Analytical Bias

Clicks, Impressions, CTR and Average Position: Definitions and Uses

Each metric has its own trap, summarised in the table below:

Average position aggregates different realities: the same page can rank at the top on mobile in France and much lower elsewhere. A featured snippet counts as position 1. The calculation, its biases and the action thresholds are covered in our guide to interpreting average position in GSC.

Understanding Data Gaps: Chart, Table and Update Delays

Several rules of how the report works explain figures that seem inconsistent (at the time of writing):

• History: the report keeps 16 months of data, which allows year-on-year comparison.
• Rows displayed: the interface shows 1,000 rows; the Search Analytics API returns up to 50,000 per request.
• Preliminary data: the most recent days are provisional and may still change.
• Anonymised queries: they do not appear in the table, whose total stays below the chart total.

To distinguish a loss of visibility from a drop in attractiveness, always read position and CTR alongside clicks.

Segmenting to Isolate Growth Levers

Dimensions to Cross: Queries, Pages, Countries, Devices and Search Appearance

The report's filters cover six dimensions:

• Query: isolate a topic, a brand or a family of intents.
• Page: track a URL, a directory or a template.
• Country: compare markets and spot a local decline.
• Device: separate mobile, desktop and tablet.
• Search appearance: measure the effect of rich results.
• Regular expression: group queries or URLs by pattern.

Crossing two dimensions turns a "site average" into a diagnosis: a page losing ground on mobile in France, for example.

Filters and Comparisons: Periods, Branded vs Non-Branded, Geographic Areas

Comparison mode and filters separate genuine changes from structural effects:

• Month-on-month: track recent actions.
• Year-on-year: neutralise seasonality.
• Branded and non-branded: stop rising branded traffic from hiding a loss on generic queries.
• Geographic area: isolate each market as soon as the site targets several, or a local clientele.

Regular Expressions: Useful Use Cases and Common Mistakes

Regex filters track query families ("price", "reviews", "best") or segment pages by URL pattern. The classic mistake is a pattern that is too broad and catches off-topic queries: test each filter over a short period and check the rows returned.

Transforming Data Into Measurable SEO Actions

Spotting High-Potential Pages: High Impressions and Low CTR

Many impressions and few clicks signal untapped potential. Compare CTR at equal positions, check the device and the SERP features displayed, then work on the title, the meta description and intent. The levers are described in our guide to optimising click-through rate.

Identifying Quick Wins: Queries in Positions 4 to 10 With Clear Intent

Exposure gains are greatest at the top of the first page: the first position captures a 27.6% click-through rate (Backlinko, 2026), according to our SEO statistics. Positions 4 to 10 with clear intent are therefore the quick wins zone. To work on it:

• filter the queries in this zone with the most impressions;
• check that the ranking page actually matches the intent;
• enrich the content and structure it further;
• strengthen internal linking to that page.

Diagnosing a Drop: Lost Queries, Cannibalisation and Seasonality

When clicks drop, first compare the period with the same period the previous year to rule out seasonality. Then check three hypotheses:

• demand has decreased: impressions down, position stable;
• visibility has fallen: position down on a specific segment;
• several pages share a query: cannibalisation.

For cannibalisation, filter the query then open the Pages tab: if several URLs capture impressions, clarify the target URL and consolidate (merge, redirect, update) rather than multiplying similar pages.

Prioritising With an Impact Logic: Traffic, Conversion and Effort

The grid below crosses the potential and the cost of each action:

Potential shows in impressions and position, business value in conversion. If you can gain without changing the page's main intent, it is a tweak; otherwise, it is a project.

Analysing by Search Type

Each search type is read separately, and Discover has its own report:

Web: Managing the Core of Organic Traffic

The "Web" search type remains central. To read it correctly:

• segment mobile and desktop;
• separate commercial and informational queries, whose CTR differs;
• read a rise in impressions without clicks as a brand awareness gain or the effect of AI Overviews.

Discover and News: Interpreting Volatility and Securing Gains

Discover and News are more volatile, because content freshness plays a key role there. Identify formats that work, read drops as normal cycles and update your reference pages.

Search Console Insights

Search Console Insights is a report built into GSC since June 2025 and works without Google Analytics, whose integration was removed at the end of 2024 (at the time of writing). Focused on content, it does not replace the Performance report for detailed analysis. Each of its cards answers one question:

This report is for a quick monthly read: spot clear changes, then switch to the Performance report to isolate a query, a device or a URL, as it hides fine segments.

AI Overviews, AI Mode and the Performance Report

AI Overviews and AI Mode rely on Google's index: a page that is not indexed cannot appear in them, so check the indexing of key pages first. Their impressions and clicks are counted in the "Web" search type, with no dedicated filter (AI Mode since June 2025, at the time of writing). This visibility cannot be isolated, but its effect shows:

• impressions rising on informational queries;
• clicks stagnating or falling at equal position;
• CTR declining with no loss of ranking.

The gap can be large: a 2.6% CTR in first position when an AI Overview is present (Squid Impact, 2025), according to our GEO statistics. If position holds, the drop comes from the results page, not your content. How to read these impressions is covered in our guide to impressions in GSC.

To be featured in them, structure your content following our guide to Generative Engine Optimization.

Combining Data With Google Analytics to Measure ROI

A click in Search Console is not a session in GA4: the two tools count neither the same unit nor at the same moment. The causes of the gap are explained in our guide to connecting Search Console and Google Analytics. GSC steers visibility, GA4 measures business contribution.

To connect SEO and ROI, start from organic landing pages and check three criteria:

• Engagement: does the page hold on to the visitor who came from the query?
• Conversion: does it generate measurable leads, sales or revenue?
• Alignment: does its promise match the query that brings visitors in?

This way, you spot pages that convert without enough visibility, and those that attract traffic with no value.

Depending on Your Situation

Depending on what your Performance report reveals, here is the resource to open next:

• Important pages have no impressions: first check that they are indexed, with the guide to indexing in GSC.
• The interface's 1,000 rows are not enough: extract more data with the guide to the Google Search Console API.
• You need to report on SEO ROI: choose the right indicators with our guide to KPIs in Google Analytics.
• Emerging queries are appearing: turn them into an editorial plan with our guide to SEO content strategy.

To track these metrics page by page in monthly dashboards and measure the effect of each optimisation, you can rely on our SEO and GEO performance reporting module.

Frequently Asked Questions About the Performance Report

How Many Months of Data Does Google Search Console Keep?

Google Search Console keeps 16 months of data in the Performance report, at the time of writing. This history allows a year-on-year comparison that neutralises seasonality. To keep a longer history, export your data regularly.

How Often Is the Data Updated?

Performance report data is not real-time: the most recent days are preliminary data that may still change. Wait a few days before analysing a recent change, and favour a monthly reading so you can decide without over-interpreting movements that are too short.

How Long Does It Take to Observe the Effect of an Optimisation?

It depends on the optimisation, crawl frequency and competition. Favour analysis windows of a few weeks, segmented by query or page. Note the date and nature of each change so you can then compare periods of the same length.

Which Comparisons to Use to Validate Impact: Before/After, Year-on-Year, Week-on-Week?

The three comparisons are complementary. Before/after measures a direct effect with consistent scope, year-on-year neutralises seasonality, and week-on-week serves operational monitoring despite its volatility. Always check whether the change comes from position, CTR or impression volume.

Do AI Overviews Count in Impressions?

Yes, AI Overviews and AI Mode are counted in the impressions and clicks of the Performance report. They are included in the "Web" search type, with no dedicated filter, AI Mode since June 2025 at the time of writing. Their effect shows indirectly: rising impressions and falling CTR at equal position.

Where Can I Find Search Console Insights?

Search Console Insights is a report built directly into Google Search Console since June 2025. It no longer requires a linked Google Analytics account, as that integration was removed at the end of 2024. It helps you quickly spot content trending up or down, before analysing in the Performance report.

Why Do My Clicks Differ From Google Analytics?

Your clicks differ from Google Analytics because the two tools measure neither the same unit nor the same moment. GSC counts a click on the results page, GA4 a session on the site, subject to consent and the configured time zone. Use each tool within its own scope rather than trying to reconcile them.