Search Console Mastery

Search Console is Google's free tool that shows how a site looks in search results. The main thing in it for an analyst is the Performance report: it shows how your search traffic changes over time, where it's coming from, and what search queries are most likely to show your site. Thresholds such as "positions 8-15" and "CTR above 10%" are common, as is the advice to build "free position tracking" from an export. Those are not in the documentation, and average position is not suited to tracking. Below: what each metric counts, how to read the chart and the table, and which tasks Google itself describes.

The basis is Google's help on the Performance report, its common tasks and the counting methodology. On linking with analytics, see the lesson on GA4; on how to open the tool, the lesson Google Search Console.

The main reports

• Performance — clicks, impressions, CTR and average position, split by queries, pages, countries, devices, search appearance and dates;
• URL Inspection. It provides information about Google's indexed version of a specific page and lets you test whether it might be indexable. You can see why a page could not be indexed, request crawling, see the rendered version and loaded resources, and tell a live test from the indexed results;
• Page indexing. The earlier name "Coverage" is out of date. The report has two duplicate statuses, and "Duplicate without user-selected canonical" is not an error;
• Core Web Vitals — speed and stability metrics by page group: the lesson on Core Web Vitals;
• Manual actions — messages about manual actions, if there are any.

Property types: domain and URL prefix

The "add all variants of the domain" advice applies only to a URL-prefix property. A Domain property includes all subdomains and multiple protocols. A URL-prefix property is for when you need to limit your data by URL path segments or by protocol. Note: the branded and non-branded filter is not available for sub-properties, such as the /blog/ folder.

What the metrics count

Data is attributed to the canonical URL of the link. This ties the report to the lesson on canonical: if Google chose a different canonical page, the numbers go to it.

Chart and table: why the totals differ

Data is aggregated either by property or by page. The chart is always by property; the table by queries, countries, devices and dates is by property too, while by pages and search appearance it is by page. Hence the discrepancies.

• By property: if the property appears twice on a results page it is one impression; two clicks on links of the same property count as one click; position is the topmost;
• By page: each unique URL counts separately, and clicks on two different URLs are two clicks.

Because of the difference CTR and average position are generally higher when aggregating by property. In Google's example of three results from one site at positions 1, 2 and 3, by property you get CTR 100% and position 1, by page CTR 33% per URL and an average position of 2.

Filters and comparison

• Search type: web (text-based or multimodal), image, video, news. Multimodal includes searches by image, Google Lens and Circle to Search;
• Time granularity: hourly, daily, weekly or monthly. Weekly and monthly smooth out weekend and holiday fluctuations;
• Queries in the table are exact matches, not case-sensitive; similar queries are combined with regular expressions;
• Branded and non-branded queries. Branded include the brand name, domain or a brand-specific product, with misspellings; non-branded don't. Data is available from 11 March 2025, for all search types, but not for sub-properties and sites with few impressions. It is a way to tell brand awareness from new audience growth; our branded vs non-branded breakdown does the same;
• Fresh data may be preliminary and change within hours; on the chart it is shown with a dotted line. Search Console keeps data for 16 months.

Tasks Google describes

1. The most popular queries: the Queries tab, sort by clicks, CTR or impressions;
2. The most popular pages: the Pages tab, sort by clicks;
3. Pages with low CTR. A low CTR can indicate that users see your page in Search results but don't think it answers their query. Select CTR as the metric, open Pages and sort ascending; if CTR is low consider updating the title and description or adjusting the content to the queries. See the lesson on the meta description;
4. Which pages were shown for a query: click a query in the table and open Pages;
5. Position history: a URL filter, the Average position metric alone, a period comparison. Google at the same time advises focusing more on trends in impressions and clicks than on position alone;
6. Significant change: sorting by the difference column shows the queries and pages that changed most.

Judging the impact of changes, and annotations

It is hard to tell whether a specific change led to growth: seasonality, news events and improvements to competing sites also contribute. Google advises checking whether the improvement maps closely in timing to your site changes, compared to similar periods in the past, and using Search Console annotations to record events on the chart. Traffic drops are in the lesson on traffic drops.

What the documentation does not support

• "Positions 8-15 mean the page is close to the top 3". There is no such threshold. It is a heuristic many people use, but priorities are better set by trends in impressions and clicks, not by position alone. Our growth-point finder highlights queries on the first page and estimates the possible gain in clicks;
• "CTR above 10% means scale it". Google names no CTR thresholds: CTR depends on result type, query and position;
• "Free position tracking via export". Average position is position averaged over impressions, and results depend on the time, place, device and recent history of the person searching. It suits trends, not "what is my position right now". See the lesson on rank tracking;
• "E-mail alerts for errors as a mandatory item". Not described in the pages cited.

Exports and comparing periods

Many reports have an export button: both chart and table data are exported, and "~" and "-" values become zeros in the download. For large exports our Search Console export works without the 1000-row limit, and the period comparison and CTR anomalies show what dropped between two periods and where click-through fell while position held.

Checklist

1. A property of the right type is verified: domain for the whole site, URL prefix for a part.
2. It is clear how impressions, clicks, CTR and average position are counted.
3. It is known why chart totals (by property) differ from the pages table (by page).
4. The search type and comparable periods are chosen for analysis.
5. Low CTR is checked through the Pages tab sorted ascending.
6. Decisions rest on trends in impressions and clicks, not one position.
7. Changes on the site are recorded with annotations.

Mistakes

• Treating average position as an exact position in results.
• Comparing chart totals and the pages table as equal.
• Taking thresholds such as "CTR above 10%" as a rule.
• Analyzing a period whose last days are still preliminary.
• Looking for the result of a change without comparing to similar past periods.

What comes next

Next is SEO KPIs: which measures to put in reporting and how to read them together.