# The Search Console performance report, column by column You look at the performance report to decide what to fix this week. Here is what each column means, the traps that skew it, and how to use it without guessing. Updated 2026-09-13 · Source: https://porteur.ai/guides/search-console-performance-report ## What the performance report counts Search Console records what happened on Google’s results, not on your site. It logs impressions, clicks and position for your pages on Google Search. It holds up to 16 months of data. The default compare view is the last 28 days against the 28 before. Change it when seasonality matters. Pick the right search type first. Web is the default. Image and Video are separate. Your numbers change a lot when you switch. Do not mix them in one judgement. Use the tabs for Queries, Pages, Countries, Devices and Search appearance to change the rows. The columns stay the same. Filters stack and carry across tabs until you clear them. ## Clicks: what a click is and common traps Clicks are visits from Google Search to your site. The click must be on a result that leads to your URL. CTR is clicks divided by impressions. - Trap: clicks fall while position rises when you lose a sitelink or a rich result. The position counts your highest link, not how many links you win. - Trap: clicks rise on brand queries after a press mention. That makes the site total look healthy while non brand pages stall. - Trap: zero clicks on a page with impressions usually means the title does not promise the answer, or the intent is wrong for your offer. Example: “yourproduct pricing” sends traffic to /pricing. The page gets impressions. If you change the title to “Pricing and plans”, watch CTR and clicks over the next 28 days, not a week with a holiday. The figures here are illustrative, not from your property. ## Impressions: when they count and when they do not An impression is counted when a link to your page appears on a results page a user loaded. For web search it counts whether or not the user scrolled to it. Discover and some carousels count only when seen. - Trap: a move from page two to page one lifts impressions sharply. Expect more impressions first, then clicks if your snippet earns them. - Trap: impressions by query never add up to site totals. Search Console omits rare and anonymised queries in the Queries table. - Trap: switching search type changes impression rules. A page can rack up web impressions and show none in Image search at all. Example: /guides/getting-started sits near the top for one query and far lower for another. It shows thousands of impressions in a month. That does not mean thousands of people saw your brand. It means results pages loaded with a link to that URL on them. These numbers are illustrative, not measured on your site. ## CTR: why the rate moves and what to fix CTR is clicks divided by impressions. Use it to judge your snippet against neighbours on the same results, not against the whole site. It is only meaningful with enough impressions. - Trap: a lower CTR after a rank rise is normal if you moved into a query with more navigational intent. The pie got bigger and harder to win. - Trap: page CTR hides query differences. Filter by one query to judge a title change. - Trap: tiny samples lie. A page with a handful of impressions and a few clicks can show a high CTR. Do not ship a rewrite on that. 1. **Find pages with impressions but poor CTR** In Pages, set a date to the last 28 days. Sort by Impressions. Scan for a low single digit CTR on URLs that matter to revenue. 2. **Open query context for one page** Click a URL, switch to Queries. Note the top three queries and their positions. 3. **Rewrite for intent and clarity** Change the title to match the query wording and the meta description to promise the next step. Example: “SaaS pricing calculator” on /pricing if that is the draw. 4. **Measure over a clean window** Wait a fresh 28 days. Use Compare to check CTR on that page for those queries only. ## Average position: how it is calculated and misread Position is the position of the highest link to your site on that results page. Average position is the mean of those over impressions. One page can average 8 while appearing at 3 and at 40. - Trap: a new low ranking query can tank your average position while your money query holds at 3. Always open the Queries tab for context. - Trap: rich results and features shift where the “first result” sits. Use position as a directional metric, not a promise of above the fold. - Trap: a URL with multiple links on one page, like a result and a sitelink, still uses the highest position for that page’s count. Example: /features/ai-assistant ranks well for “ai assistant for sales” and much lower for “sales ai”. With similar impression counts, the average position sits between those two. Do not read that as a fixed page number. Read it as one strong query and one weak query on the same URL. You might need a new page for the second query. ## Queries vs pages: why tables do not match The Queries table omits rare and anonymised queries, so its totals are lower than the site or page totals. This is by design. You still get the full site and page numbers. - Use Pages to decide which URLs to fix. Use Queries inside one page to decide how to rewrite that page. - Expect sudden drops or spikes on Queries after a product rename, a PR hit, or a new feature launch. Long tail queries shift a lot. - If a query matters and is missing, search for it on Google and check which URL ranks. You may be splitting intent across two pages. Example: Pages shows /blog/how-we-price with thousands of impressions. Queries sums to fewer. That gap is the hidden long tail. If you need more detail, add content for the themes you can see rather than chasing every missing keyword. These figures are illustrative only. ## Filters that change meaning: date, search type, device, country Set the date first. The default compare is last 28 days vs previous 28. If you ship often, anchor analyses on your release dates. For seasonal terms, compare the same period last year within the 16 month limit. - Search type: Web, Image, Video, News. Each is a separate dataset. A page can do well on Web and show nothing in Image. - Device: Desktop, Mobile, Tablet. Some snippets only show on one device. If CTR collapses on mobile only, fix the title truncation and first sentence. - Country: market intent differs. Your /pricing might win the UK and lose the US. Localise the copy and currency if the query is transactional. Do not add too many filters at once. You will hide the sample you need to see patterns. Start broad, then narrow by one dimension to confirm a hunch. ## Search appearance and result types Search appearance groups special result types. These can include things like product results, FAQ, or other rich displays as of 2026. A change here can swing CTR without a rank change. - Trap: losing a rich result removes visual hooks. CTR falls even if average position holds. - Trap: mixing search appearances in one view hides the cause. Filter to one appearance and chart CTR and impressions. - Action: click a search appearance, then switch to Pages. Fix the marked URLs to restore eligibility or align the snippet with intent. Example: /docs/api shows in a rich result one month, then plain blue links the next. CTR halves at the same position. The fix is in the page’s structured data or in how the snippet answers the query, not in links to the page. ## The compare view without false wins The default comparison is the last 28 days against the 28 before. This is useful for steady sites. It lies for launches and holidays. 1. **Anchor on events you control** Mark ship dates, press hits and outages on your own calendar. Read page and query changes around those dates. 2. **Use like for like ranges** If a term is seasonal, compare 28 days this year against the same 28 last year within the 16 month window. 3. **Hold filters steady** Pick Web only, or Image only. Do not compare Web this month with Image last month. 4. **Read deltas by column** Clicks up and CTR down can both be good if impressions jumped. Average position worse and clicks up can be good if you added new long tail. ## From reading to action: common fixes with the columns - Striking distance: filter a key page to queries at positions 8 to 15. Add a clear H1 and an answer paragraph. Add one internal link from /pricing or your home page using the query wording. - Pages shown but never clicked: sort Pages by impressions, then find a very low CTR. Ship a title rewrite that matches the top query. Example: change “Features” to “Sales pipeline automation features”. - Average position masking a winner: a page averages in the mid range but hides a query at 3. Split the weaker query into a new page. Link both ways. - Search type gaps: Image search shows impressions for /gallery but no clicks. Add descriptive file names and alt text. Add links from the copy above the fold. - Device mismatch: CTR on mobile slumps. Shorten titles to about 60 characters and front load the hook. Move the answer into the first sentence. - Country intent: the US shows low CTR on /pricing. Add USD first and an anchor link to Plans. Keep the price in the meta description if policy allows. A fixed page looks like this: /guides/getting-started shows on page one for “get started with yourproduct”, CTR in a healthy range, and the first paragraph answers the task in one line. Internal links point to it with the same wording. > If you see many URLs excluded in Page indexing as “Alternate page with proper canonical tag”, that means Google agreed with a canonical tag pointing to a different URL. It is only a problem when the canonical is wrong. Fix the canonical before chasing performance for those URLs. ## Questions ### Why do impressions rise without clicks? You likely moved up the page or onto page one, so you are seen more. Your snippet, not your rank, is now the bottleneck. Rewrite titles and meta descriptions to match the top query on that page. Watch CTR and clicks over the next 28 days. ### Why does average position worsen while clicks go up? You added impressions from new queries that rank lower. The mean position drops, but you earned more total traffic. Check the Queries tab for the page. Keep the gains and consider a new page for the weaker query if it has value. ### Why do query totals not match page totals in Search Console? The query dimension omits rare and anonymised queries. Pages and site totals include all impressions and clicks. Use the Pages tab for decisions on which URLs to fix. Use Queries for per page rewrites. ### What is a good CTR in Google Search? There is no single good rate. Judge CTR by query and position, not by site average. A navigational brand query at position 1 will dwarf a generic query at position 5. Use a 28 day window and enough impressions to avoid noise. ### How do I view performance for AI related result types? As of 2026, if Google supports a special appearance for AI driven results in your property, it shows under Search appearance. Filter by that appearance to see clicks, impressions, CTR and position. If you do not see it, your site may not have eligible impressions yet. ### How do I export this to Google Sheets? Use the Export button at the top right of the report. Pick Google Sheets or CSV. The export respects your current filters, date range and tab. Keep a note of the filters in the sheet title so you remember what you pulled. ## Read next - [How to use Google Search Console in ten minutes a week](https://porteur.ai/guides/how-to-use-google-search-console): A quick weekly routine: set four filters, compare 28 days, check pages then queries, and fix three findings, without getting lost in noise. - [What an impression means in Search Console, and what it does not](https://porteur.ai/guides/what-is-an-impression-in-search-console): An impression is a counted view of your link on a Google results page. Here is what it includes, what it omits, and what to do with it. - [What is a good click-through rate in search, by position](https://porteur.ai/guides/what-is-a-good-click-through-rate-in-search): CTR depends on position, query intent and the results page. Learn how to benchmark your own CTR in Search Console and what to fix on low-CTR pages. - [Alternate page with proper canonical tag: what Search Console means](https://porteur.ai/guides/alternate-page-with-proper-canonical-tag): What this Search Console status means, when to ignore it, when it hides the wrong page, and the exact checks and fixes to set the right canonical. - [Canonical tags: what they do and the mistakes that cost rankings](https://porteur.ai/guides/canonical-tag): What a canonical tag does, when to use one, the mistakes that cost rankings, and how to check and fix canonicals on a small site. - [What a good SEO report contains, with an example](https://porteur.ai/guides/seo-report-example): See what a good SEO report contains, how it drives decisions, and a worked example on a product page you can copy for your next report. - [The Search Console API: pull your own search data](https://porteur.ai/guides/search-console-api): What the Google Search Console API gives you, how to authorise it, the request you send, quotas, BigQuery export, and three automations to ship. Paste your home page URL to get a free 30 second check of your searches, rivals and pages with three findings you can act on before you open Search Console. Free check: https://porteur.ai/