Read your Search Performance

Open Sidebar → Measure → Performance and select the Search Console tab.

The Four Metrics

Each compares against the previous period of equal length.

Reading Them Together

The individual numbers matter less than the pattern between them. Four common shapes:

Impressions up, clicks flat. You are being shown more and chosen no more often. Usually a title and description problem, and the fix is in the SEO tab of the relevant articles.

Position improving, clicks flat. You may be moving up in places nobody clicks, for example from 40 to 25. Real change happens crossing into the top few results.

Clicks up, position flat. Often seasonal demand, or you gained a result feature. Worth understanding before you attribute it to your work.

High CTR, low impressions. You are good at winning the clicks you are shown for, and you are not shown enough. This is a demand problem, and the answer is more content on adjacent topics rather than optimising what you have.

Top Queries

The most useful panel on the page.

These are the actual phrases people used. They routinely differ from the keywords you targeted, and the gap is where your next articles come from.

Look for:

Queries you rank for and did not write about. Free direction. Search demand you already partly satisfy.

Queries with impressions and no clicks. You are visible and unconvincing. Check what ranks above you.

Question phrasings. Direct candidates for FAQ sections, which also feed the GEO signals in your Content Score.

The panel notes that "Query metrics are Search Console dimensional detail and may not add up to canonical property totals." That is a Google characteristic, not a discrepancy to chase.

There is a search box for query text, and a Trend metric filter offering Clicks, Impressions, CTR, and Average position.

Top Pages

Which content earns the traffic.

The pattern worth acting on is a page with strong impressions and weak CTR. That is a page already ranked where the title and description are not winning the click, and rewriting metadata is far cheaper than writing a new article.

Search page titles or paths with the box provided.

Clicks by Country and Devices

Clicks by country groups search clicks by visitor country. Useful for confirming you are reaching the markets in your Brand Profile, and for spotting demand somewhere you did not expect.

Devices groups clicks by device category. If mobile dominates and your mobile experience is poor, that is worth more than any content change.

A Caveat on Filtering

Filtered totals can undercount. Google withholds low-volume country and device data for privacy, and the page says so where it applies.

Treat filtered figures as indicative and unfiltered property totals as authoritative.

Choosing a Period

1d and 7d are noisy. Day-to-day movement in search is mostly randomness.

30d is the sensible default for spotting change.

90d is where trends become real. Use it before concluding anything about whether your content is working.

Remember every range compares to the previous equivalent period, so a 30d view compares against the 30 days before it.

The Three Day Lag

Search Console reports about three days behind. The last few days of any window will look artificially low, and that is Google, not NarraSEO.

Do not judge a newly published article on search data for two weeks. There is nothing to see yet.

Notes

Sync now pulls the latest available data. It cannot retrieve data Google has not published.

Metrics show a dash where there is no data, rather than a zero.

Where there is no earlier period to compare against you see "No prior baseline" instead of a change figure.

Turning This into Work

The honest short version:

1. Top queries for what to write next.
2. Top pages with weak CTR for metadata to rewrite, which is the cheapest win available.
3. 90d trend for whether the whole effort is working.

Everything else is context.

Related

What the Performance page shows covers the page as a whole. Editing your SEO title, description, and schema is where you act on a CTR problem.