Search Console guide

How to find pages losing traffic in Search Console

Direct answer

To find pages losing traffic in Search Console, compare the last 28 days with the previous 28 in Performance, open the Pages tab and sort by the click difference, biggest drop first. Then read each loser's impressions and position: fewer impressions points to demand or ranking, a lower CTR points to the listing itself. Rankgaroo lets an AI pull both periods by page and build that ranked list in one request.

Updated 2026-10-25

On this page

Do it by hand Ask your AI instead Worked example Limits to know Questions Related guides

You know some pages used to bring clicks and now bring fewer. Search Console lists every page, but it will not tell you which ones slipped unless you compare two date ranges yourself.

A useful loser list needs more than click counts. A page can lose clicks because it ranks lower, because fewer people search the topic, or because your other pages now take the same searches.

Do it by hand in Search Console

In Search Console, open Performance → Search results.
Open the date filter, choose the Compare tab and compare the last 28 days with the previous period.
Select the Pages tab and tick Clicks, Impressions, CTR and Position so all four are columns.
Sort by the clicks difference column so the biggest drops come first.
For the top 10 to 20 rows, read the impressions and position change next to the click change, and note which of the three moved.
Click a loser to filter on it, then open the Queries tab to see which searches stopped sending clicks.

You read four metrics across separate columns for each page, and finding the queries behind every loser means filtering and checking one page at a time.

Ask your AI instead

With Rankgaroo connected to Claude, ChatGPT or Cursor, paste a question like this:

Prompt

Using my Search Console data for sc-domain:example.com, compare 2026-09-25 to 2026-10-22 with 2026-08-28 to 2026-09-24 by page. List the 10 pages that lost the most clicks. For each, show impressions and average position before and after, say whether the loss looks like fewer impressions, a lower position or a lower CTR, and flag any pair of pages that seem to trade clicks.

What the AI calls

gsc_search_analytics(siteUrl: "sc-domain:example.com", startDate: "2026-09-25", endDate: "2026-10-22", dimensions: ["page"], rowLimit: 1000)
then the same call for 2026-08-28 to 2026-09-24. The AI joins the two lists by URL and subtracts clicks.
gsc_search_analytics(..., dimensions: ["query"], filters: [{dimension: "page", operator: "equals", expression: "<page URL>"}], rowLimit: 100)
For each loser. The AI lists the queries behind the loss.

Worked example

Page	Clicks before	Clicks now	Change	Impressions before	Impressions now	Position before	Position now
/templates/invoice-template/	640	572	-10.6%	36,800	38,900	8.1	8.3
/ja/templates/timesheet-template/	212	153	-27.8%	6,940	4,880	6.8	6.6
/blog/freelance-tax-deductions/	190	131	-31.1%	24,600	17,200	9.6	9.6
/templates/timesheet-template/	300	254	-15.3%	21,500	14,300	7.4	7.0
/es/plantillas/factura-autonomo/	96	55	-42.7%	2,150	1,370	5.9	6.1
/blog/invoice-payment-terms/	118	84	-28.8%	15,300	15,100	9.8	10.9
/blog/time-tracking-for-agencies/	64	30	-53.1%	5,480	3,010	12.4	12.5
/blog/how-to-write-a-quote/	87	54	-37.9%	9,300	9,150	8.8	8.9
/templates/quote-template/	75	45	-40.0%	11,900	8,100	8.6	8.4
/pricing/	58	32	-44.8%	3,400	3,520	2.6	2.8

Example site · Search Console · 2026-09-25 to 2026-10-22 vs 2026-08-28 to 2026-09-24. Sample data from a fictional site.

What the answer says

The largest single loss is /templates/invoice-template/ at 68 clicks (640 to 572), but its impressions rose and its position barely moved, and a sibling page, /templates/simple-invoice-template/, gained 63 clicks, so check whether the two split the same searches. Most other losers show a different shape: impressions fell by about 30% or more with position steady, as on /blog/freelance-tax-deductions/ (24,600 to 17,200 impressions, position 9.6 in both periods), which points to fewer searches or fewer queries rather than a ranking loss. The data alone cannot confirm either reading.

Limits to know

Search Console returns up to 1,000 rows per call. On a site with thousands of URLs, the list covers the busiest pages and can miss losers in the long tail.
A page with no rows in one period counts as zero there, so a page that dropped out of results and a page that was never indexed look the same.
Low-volume pages swing a lot. A loss of 10 clicks on a page that earned 30 is often noise, so read percentages next to raw clicks.
The data ends about 3 days before today, and Rankgaroo cannot inspect URLs or check indexing, so a lost page still needs a manual look.

Questions

How do I see which pages lost traffic in Search Console?

Open Performance → Search results, set Compare to the last 28 days against the previous period, select the Pages tab and sort by the clicks difference. The pages at the top of the list lost the most clicks. Read impressions and position beside each one to see what changed.

Why would a page lose clicks if its position did not change?

A page can lose clicks at the same position when fewer people search the topic, when the listing's CTR drops, or when another page on your site takes some of the same queries. Search Console shows impressions, CTR and position, so you can test the first two. Anything on the results page itself needs a manual look.

How big a click drop on a page counts as a real loss?

Compare the drop with the site's normal swings between periods. A page with hundreds of clicks losing 30% or more deserves a look; a page with 20 clicks losing 8 usually does not. Persisting across several periods matters more than any single drop.

Can Rankgaroo build the list of losing pages for me?

Yes, as a one-time pull. The AI calls Rankgaroo for page data in both periods, joins and sorts it, and can pull the queries for each loser. Rankgaroo stores no history and sends no alerts, so you ask each time you want a fresh list.