Search Console Performance Report: 5 Essential Metrics

Hey! If you have ever stared at a spreadsheet of queries in Search Console and thought "okay, now what do I actually do with this," this one is for you. We are going to turn that raw table into a short list of pages you can fix this week.

Picture this: a page you wrote months ago is sitting on page two of Google, quietly collecting thousands of impressions and almost no clicks. You would never notice it by scrolling your own site. But a Search Console Performance Report would show it to you in about thirty seconds, if you know where to look.

Open Search Console and click Performance in the left sidebar, and you land on the Search results view. This is the only place Google hands you real data about how people found your site in organic search, straight from its own logs rather than a third party estimate.

The Search Console Performance Report is built around one table and one chart, both of which respond live to whatever filters and date range you set at the top of the page. Every row in that table represents either a query, a page, a country, a device, or a search appearance type, depending on which tab you have selected.

Before you dig into the technique in this guide, it helps to have your property configured properly so the data you are reading is complete and accurate. Our guide to setting up Search Console walks through verification and sitemap submission if you have not done that yet.

Every row in the Performance report reduces down to four measurements, and Google defines them precisely rather than leaving them open to interpretation. Clicks are, in Google's own words, how often someone clicked a link from Google to your site. Impressions count how often a link to your site was seen in results, even if nobody scrolled far enough to notice it, according to Search Console Help.

Click through rate is simply clicks divided by impressions, expressed as a percentage, and average position is where your link typically sat on the results page, counted from the top down. None of these four numbers tells the full story on its own, which is exactly why the Search Console Performance Report pairs them together in one view.

Metric	What It Measures	What To Watch For
Clicks	Visits that started from a Google result	Flat clicks despite rising impressions
Impressions	Times your link appeared, even unscrolled	High volume with little traffic to show
Click through rate	Clicks as a share of impressions	A rate far below the average for that position
Position	Average rank across all matching queries	Pages hovering just below the first page

It is tempting to treat position as the only metric that matters, but click through rate can swing wildly between two pages sitting at the same spot. A benchmark study cited by Semrush puts average click through rate near 39.8 percent at position one, dropping to 18.7 percent at position two and 10.2 percent at position three, and the same research notes that only about 0.63 percent of searchers ever click something on page two at all.

The landscape has also shifted recently because of AI generated summaries appearing above the organic results. Ahrefs analyzed three hundred thousand keywords and found position one click through rate fell by roughly 34.5 percent when an AI Overview appeared on the page, and a separate study of seven hundred thousand keywords by Amsive found informational queries without a brand name lost around 20 percent of their clicks, as reported by Search Engine Land.

That shift matters for how you read your own Search Console Performance Report. A page that used to convert impressions into clicks reliably at position three might now show a lower click through rate simply because an AI summary sits above it, not because your content got worse.

Here is the actual quick win technique at the center of this guide. Open the Queries tab in your Search Console Performance Report, sort the table by impressions from highest to lowest, and scan for rows with an average position somewhere between four and fifteen. Those queries already rank, Google already trusts the page enough to show it often, but it has not quite broken into the top three where most of the clicks land.

Search Engine Journal describes this group as striking distance keywords, originally defined as positions four through twenty, though the article notes the range is meant to be adjusted rather than treated as a fixed rule. A more recent playbook from GSC Wizard narrows the practical sweet spot to positions five through fifteen, arguing that pushing a page from position fifteen to position three is a far easier lift than pushing it from position five.

That same playbook recommends ignoring any query with fewer than fifty impressions across a ninety day window, since smaller samples are too noisy to act on with confidence. One worked example it cites shows a query sitting at a 1.3 percent click through rate despite thirty four hundred impressions in ninety days, which is exactly the kind of row worth your time.

Once you have your short list, cross check the surviving queries against your existing keyword research to confirm search intent before you touch the page. Our comparison of keyword research tools is a useful second opinion if a query looks promising but unfamiliar.

The dimension tabs along the top of the Search Console Performance Report, covering queries, pages, countries, devices, search appearance, and dates, are not meant to be used one at a time. According to Search Console Help, you can apply several filters together so the table only shows rows that match every condition at once.

A practical combination for the four to fifteen technique looks like this: filter to a single page, add a country filter for your main market, then switch to the Queries tab to see exactly which search terms are dragging that page's average position down. Flip to the Devices tab afterward, since a page that performs fine on desktop but badly on mobile often has a technical cause rather than a content one.

Before you act on any list of queries, filter out your own brand terms so they do not crowd out the generic opportunities you are actually looking for. Search Console has a built in way to do this, which our guide to the branded queries filter explains step by step, and a similar manual approach using combined filters is described by Resignal, which recommends a wide six month window when exporting data to spot topic patterns across many queries at once.

Most of the confusion around this report comes from treating it like a ranking tracker instead of a log of real search behavior. It is neither wrong nor broken when a number looks odd, it is usually just being measured differently than you expect.

If a page is missing from the Search Console Performance Report entirely, that is an indexing question, not a performance question, and pulling it apart with filters will not help. Our walkthrough of the URL Inspection tool is the right next step for that specific problem.

A Search Console Performance Report is honest but incomplete, and knowing its edges saves you from chasing phantom problems. Google retains this data for only sixteen months, after which older rows are permanently removed from the interface, as noted by SEO Testing.

The numbers you see are also not instant. Search Console itself warns of a lag between when activity happens and when it shows up in your account, and in practice that gap typically runs about two to three days, per the same source and Google's own About Search Console data documentation. Rows for very rare, low volume queries are also omitted from the table to protect searcher privacy, which mostly affects long tail terms rather than the mid volume rows this guide focuses on.

Pair the Performance report with your crawl data for a fuller picture, since a page can look fine in one and still have a problem the other would catch. Our guide to the Crawl Stats report covers the crawling side of that equation.