Search Console Performance Report: Query and Page Analysis

The Search Console performance report is the closest thing you have to a first-party record of how your site appears in Google Search. It tells you how often you were shown, how often you were clicked, which queries and pages were involved and where the audience sat. Yet most people open it, glance at the line chart, and close it. The value is in the cuts you make: by query, by page, by country, by device and across two time periods side by side.

This guide is about reading the report itself. We cover what each metric means, how the date and comparison controls change the story, how to find opportunity queries and separate brand from non-brand demand, and how to read the report alongside indexing data. It closes with a monthly workflow you can repeat. Tactics for lifting click-through rate and campaign tagging with UTM parameters live in their own articles, which we link where they fit.

What the Search Console Performance Report Shows and Does Not Show

According to Google's help documentation, the report summarises your visibility in Google Search results, and you can slice it by queries, pages, countries, devices, search appearance and dates. The default view covers the past three months.

Know the boundaries before you rely on it.

• It shows how people who found you in Google Search behaved at the results page: what they saw and what they clicked.
• It does not show what happened after the click. Time on site, conversions and revenue belong to your analytics tool.
• It does not show traffic from other search engines. This is Google data only.
• It does not promise completeness. Some data is summarised, and some queries may be withheld for privacy reasons. Check the help documentation on how the data is processed before drawing fine conclusions from rare queries.

Freshness is another boundary. Google says the newest data can be preliminary, meaning it is still being collected and may change over the next few hours. Preliminary values appear on a dotted line. If you make decisions on yesterday's numbers, you may be reacting to a figure that has not settled.

Open the report with a question already in mind. "Did traffic fall?" is a question. "Let me have a look at the data" is a habit. Without a question the report becomes a wall of numbers, and each section below shows which view answers which question.

The Metrics: Clicks, Impressions, CTR and Average Position

The four metrics look simple, which is exactly why they get misread.

Clicks count the times a user clicked through to your site from Google Search results. Impressions count how many times your site appeared in results. CTR is clicks divided by impressions. Average position, in the chart, reflects the topmost result your site had across the whole property; in the table, it is the average position for the query, page or other grouping in that row.

Chart and table are aggregated differently, and that explains many apparent contradictions. Google's documentation states that chart data is aggregated by property, so two results from your site for one query count as one impression. Table data grouped by queries, countries, devices or dates is also aggregated by property, while table data grouped by pages or search appearance is aggregated by page. As a result, chart totals and table totals can differ. That is a feature of the method, not a bug in your data.

Keep three rules in mind when you read average position:

1. It is an average across many users and searches, and results vary by time, place, device and search history.
2. For one query it is informative. For a whole site it blends unrelated queries and says little.
3. Small wobbles are normal. Large, persistent shifts deserve investigation.

When CTR looks weak, the usual suspects are how the result looks and what surrounds it on the page. Methods for improving it have their own write-up in how to improve organic CTR with Search Console. In this article we only care about how CTR appears in the report and what it can and cannot tell you.

Date Ranges and Comparison Settings

Choosing the right comparison is the shortest route to the right conclusion. The report lets you pick a date range and put two periods next to each other. Google's guide to debugging traffic drops mentions comparing the last three months with the previous period, comparing year over year and widening the view to the last 16 months. The wide view is the quickest way to see whether a dip is a recurring seasonal pattern or something new.

Common mistakes with dates:

• Comparing a partial period with a full one. Mid-month, "this month versus last month" always shows a decline. Compare periods of equal length.
• Ignoring weekday patterns. In short windows the mix of weekdays and weekends skews results. Weekly aggregation smooths daily noise.
• Forgetting seasonality. A range containing a holiday or campaign cannot be fairly compared with an ordinary month. Year-over-year is fairer.
• Treating preliminary data as final. Compare completed days.

The documentation also explains time granularity. Hourly points are available only in the 24-hour view, where daily is not offered, and weekly and monthly views smooth daily fluctuation. On a volatile site, switching to weekly makes the trend far easier to see.

A habit worth adopting is to run two comparisons for every analysis. First, against the previous period to see short-term change. Second, against the same period last year to see the seasonal backdrop. If both point the same way, trust the signal. If they disagree, suspect seasonality or an outside factor before blaming your own changes.

Query and Page Breakdowns

The report's real strength is that the same data can be cut along different dimensions. Queries and pages are the two core cuts, and they pay off most when read separately and then together.

The query view shows the phrases people used when they found you. It surfaces terms you never targeted, spelling variants and question-style searches, and it is a steady source of content ideas.

The pages view shows which URLs earned impressions and clicks. For loss analysis, Google suggests sorting the pages table by the change in clicks and asking whether the loss is site-wide, confined to a group of pages or limited to one. That simple step turns a vague "traffic is down" into a concrete list.

The insight appears when you combine the two. Filter to a page and look at its queries, or filter to a query and look at which pages appeared. The second move reveals whether more than one of your pages competes for the same search. The meaning of that pattern and how to confirm it are explained in keyword cannibalization and how to detect it. A later section shows how the report presents the signal.

A reliable reading order looks like this:

1. View the site-wide chart first to understand direction.
2. Open the pages table to find the pages that changed most.
3. Pick one page, switch to its queries and see which ones moved.
4. Investigate the cause elsewhere: index status, content edits, changes in the results page.

The order matters. See the shape, then the page, then the query.

Country, Device and Search Appearance Filters

Overall numbers often hide one segment's movement inside another. Filters are how you locate a change, and knowing which to apply saves time.

Google's help article lists search types such as web, image, video and news, and its debugging guide advises examining each type separately to find where a drop occurred. On multilingual sites the country filter is especially useful. Reading the Turkish edition in Turkey and the English edition elsewhere shows whether each language version is working in its intended market.

Two filter mistakes are common. One is stacking so many filters that the data shrinks to a tiny, noisy set. The other is forgetting a filter is active and drawing conclusions about the whole site. Reset filters after each analysis, or save and name the views you use regularly.

Filters combined with comparisons can also mislead through small samples. In a segment with few impressions, a difference of two clicks looks like a dramatic swing in relative terms. Judge the absolute numbers before reacting to the shape.

Finding Opportunity Queries: High Impressions, Low CTR

An opportunity query is one where Google already shows you but searchers do not click enough, or where your page hovers at the start of the second results page. Improving visibility you already have often costs less than creating new content.

To select candidates, list queries with high impressions, a position in the lower half of page one or the top of page two, and a CTR that looks low for that position. Then ask three questions about each.

• Intent match. Does the page truly satisfy the need behind the query?
• Appearance. Do the title and description match the query and earn the click?
• Depth. Does the page cover the topic as completely as competing results, or is a section missing?

If the list contains question-style queries, a page with a short, clear answer block might compete for the highlighted answer box at the top of results. How that is achieved is covered in how to win featured snippets. Here the point is simply that high-impression question queries are usually the first candidates to check.

Set your priorities deliberately: queries close to your business goals first, then the highest impressions, and last the interesting but irrelevant. High impressions do not always carry business value. An informational query unrelated to your offer can pull you in the wrong direction.

Also avoid fixating on a single query. Consider all the queries a page receives together, because one page-level improvement frequently helps several of them at once.

Separating Brand and Non-Brand Queries

People who type your brand name already know you. Brand queries usually bring strong CTR and top positions, which flatters the totals. Non-brand queries show whether your content is reaching new audiences. Mix them and you lose sight of what SEO is actually contributing.

To separate them, filter queries by a pattern that includes your brand name and common spelling variants, and save one view that includes and one that excludes it. Remember misspellings, abbreviations and product names.

The two views let you read different things:

• Rising non-brand impressions suggest your content is appearing for new phrases.
• Rising brand queries can reflect awareness or campaigns, so do not credit them to SEO work.
• If non-brand clicks fall while brand clicks stay flat, a calm total hides a real loss.

This split is also useful when you assess outside activity such as sponsored articles. A rise in brand searches after a placement may be an awareness effect independent of the link itself. For tracking such work with campaign parameters, use the approach in sponsored post reporting with UTM, GA4 and Search Console. This article stays with the Search Console side of the brand split.

Reading Cannibalization Signals

Cannibalization means several pages on your site compete for the same query. In the report you can trace it by combining a query filter with the pages view: choose one query, open the pages tab, and if more than one URL earns impressions you have found a candidate.

Read the signal with care, because two pages appearing for a query is not automatically a problem. Google sometimes shows two results from the same site. These patterns are more telling:

1. The pages swap positions over time and neither holds a stable rank.
2. Both pages serve the same intent, and one looks like a weaker copy of the other.
3. Impressions are split across two URLs, and the combined result is weaker than a single strong page would achieve.

Typical remedies include merging content, differentiating one page's intent or using internal links to promote the preferred page. Diagnose first, and decide second.

There is also a signal in silence. If a page you published receives no impressions at all, check whether it is indexed before suspecting cannibalization. A page missing from the index cannot show up in the performance report. For a structured way to check, see how to diagnose pages not indexed by Google.

No single measurement proves cannibalization. Query, page, time and intent need to be read together, and the report should be treated as a generator of hypotheses that you then verify.

Reading the Report Together With Indexing Data

The performance report answers how visible you were. The indexing reports answer which of your pages Google knows about. Reading both resolves many puzzles.

Some common pairings:

• A page has no impressions and the indexing report says it was not indexed. The problem is indexing, not visibility.
• Impressions dropped sharply on one date and indexing errors rose on the same date. A technical event such as a faulty noindex or a server issue is a likely cause.
• Impressions held steady while clicks fell. Look at the result's appearance or at a change in what the query means rather than at indexing.

Google's traffic drop guide supports this approach. If both impressions and clicks fell, look into the usual causes. If impressions held but clicks fell, review titles, snippets and competing rich results. It also recommends the page indexing report for site-wide problems and URL Inspection for individual pages.

To present all of this in a tidy format, how to create an SEO report with a monthly template shows how to combine performance and index data on one page.

One more caution. The performance report only describes results that were shown. A page that is indexed but never displayed will not appear in the table at all. If you compare the report's page list with a full inventory of your site, you will also catch the pages that never appear.

A Monthly Analysis Workflow

Rather than opening the report at random, follow the same sequence each month. A routine can be finished within an hour at the start of the month and makes your decisions comparable over time.

1. Data integrity. Check for any notice of data issues or logging errors, then select a completed period.
2. Overall direction. Compare last month with the one before, then with the same month last year.
3. Segment. Split by search type, country and device to find where the movement happened.
4. Pages and queries. List the biggest winners and losers of both.
5. Brand split. Compare brand and non-brand views.
6. Opportunities. Choose no more than five queries with high impressions and low clicks.
7. Index check. Confirm that new and critical pages are indexed.
8. Decisions and tasks. Give every finding an owner, an action and a review date.

Many of the actions that come out of the opportunity list are updates to existing articles. Deciding between a light edit and a deep rewrite is covered in content refresh: updating old blog posts. Because Google suggests waiting a few weeks before judging whether a change helped, put the review date several weeks ahead.

Keep a change log as well. If you do not know which page was edited on which date, you cannot connect next month's movement to that edit. The simplest log has three columns: date, page and what was done.

Sample Template: Monthly Performance Analysis Table

This table is a hypothetical example. The observations are invented purely to demonstrate the format, so fill it with figures from your own report. The imagined case is a small software consultancy running its month-end review.

As you adapt it, write the likely cause column as a hypothesis to verify, not a verdict. The review date column is what turns a report into a task list, so never leave it blank.

Frequently Asked Questions

Is Search Console data real time?

Not fully. Google's help documentation says the newest data can be preliminary, still being collected and liable to change within hours, and the chart shows it as a dotted line. Base decisions on completed days. The hourly view covers only the last 24 hours and includes preliminary values.

How is average position calculated?

The documentation says the chart shows the average position of the topmost result for your site across the property, while the table shows the average position for the query or page in each row. It is an average across many users and searches, so it works better for a single query than for a whole site.

Why do some queries not appear in the report?

Data is summarised and some queries may be withheld for privacy reasons. Chart and table are also aggregated at different levels, so totals may not match. For limits and details, read the data explanations in the Search Console help documentation rather than assuming, and avoid drawing fine conclusions from rare queries.

What should you check first when CTR is low?

Check impressions and position first. If impressions are high and position is reasonable, the issue is usually the title, description or how the result looks. Then confirm the query intent matches the page and see whether competing results use richer formats. Improvement tactics are a separate topic from reading the report.

How do you filter out brand queries?

Filter queries with a pattern that includes your brand name, spelling variants and product names. Save one view that contains those queries and another that excludes them, so you can read brand and non-brand performance separately. Remember misspellings and abbreviations, since missing them leaves brand traffic inside the non-brand view.

Why does performance data not match Analytics?

The two tools measure different things. Search Console counts visibility and clicks in Google Search, while analytics counts sessions on your site. Definitions, aggregation levels, time slicing and browser conditions differ. Do not expect an exact match. Check whether direction and proportions stay consistent over time.

Conclusion and Next Step

Reading the performance report well is less about looking at more numbers and more about asking the right question with the right cut. Know what each metric means, remember that chart and table aggregate differently, treat preliminary data cautiously, compare against two periods, separate brand from non-brand and pick opportunity queries with business goals in mind. Repeat the same sequence monthly and the report turns from a dashboard into a decision process.

For this month, choose one page, compare the last three months against last year and write the five queries that changed most into the table above. If the opportunity list turns up articles you would rather commission, explore the content writing service, where the order flow on Akua Adforce begins with your choice, continues into the basket and payment, needs the seller to approve it, closes with a delivery notice and leaves a route to dispute the result.