Google Search Console Report: Find Declining Pages and CTR Opportunities

Two lists answer most of a monthly Search Console review: pages that lost clicks, and queries that get seen but not clicked. Both are a comparison of two windows.

By Ed · Published October 25, 2026 · 7 min read

Twenty-five years starting companies, consulting, owning a multi-million-dollar P&L and leading business intelligence for large enterprises.

To find declining pages in Search Console, compare two equal windows (this month against last month and last year), list pages by change in clicks, and then check whether impressions fell too. Pages where both fell lost visibility; pages where only clicks fell are being shown and chosen less. To find CTR opportunities, list the queries with many impressions and a low click-through rate relative to the rest of your site. Below is the method, a worked example with illustrative numbers, and how to ask the same questions in Quiriz.

Two windows, chosen carefully

Nearly every false alarm in a Search Console review comes from the comparison, not the data. Three rules remove most of them.

Use complete months. Search Console days are Pacific Time and Google adds the newest two or three days once they are final. A month that ended yesterday is not complete.
Compare with last year as well as last month. A page about gift guides will fall in January. Search Console keeps 16 months, so last year's month is always there.
Mind the weekday mix. A 30-day month with five weekends against a 31-day month with four will move weekday-heavy traffic on its own. Look at clicks per day when the lengths differ.

Finding declining pages

Take clicks per page for each window, subtract, and sort by the loss. Two filters make the list usable. First, require a minimum number of clicks in the earlier window: a page that went from 6 clicks to 2 is arithmetic, not a trend. Second, keep the absolute change next to the percentage, because a 60 percent fall on a small page and a 10 percent fall on your best page are different priorities.

Then split the list by what happened to impressions. This is the diagnostic that saves the most time:

Impressions	Clicks	What it usually means	First check
Down	Down	The page is shown less often	Position, indexing, a template or robots change
Flat	Down	Shown as often, chosen less	Title, snippet, a new result above it
Up	Flat or down	Ranking for more queries, mostly low on the page	Usually not a problem; check the main queries

An example, with made-up numbers. A product comparison page had 1,200 clicks last month and 700 this month, a loss of 500. Impressions went from 40,000 to 39,000, so it is still being shown. Average position moved from 4.1 to 5.3. The page is shown but chosen less, and it slid one position. That points to the title and to whoever now ranks above it, not to indexing.

Finding CTR opportunities

The second list works on queries. Sort queries by impressions, then read the CTR and position next to each. The rows to pull out are those ranking on the first page with a CTR clearly below your own pages at a similar position. Do not borrow an industry CTR curve: search results differ by query, and your own site is the honest benchmark.

A worked example, again illustrative. A query for a pricing question has 30,000 impressions at average position 5.0 and a 1.2 percent CTR, while your other pages ranking around position 5 earn about 4 percent. If this query reached 3 percent, that would be roughly 540 more clicks a month on the same ranking. The action is a title rewrite, not new content.

Be careful with CTR on its own. A low CTR at position 18 is expected. A low CTR on a query where you rank first and Google shows an answer box above you may not be fixable. Always read CTR with position.

The traps

Brand queries hide the story. A branded search lift can mask non-branded decline. Look at the two separately where you can.
Page cutoffs look like drops. Search Console data is limited to the top pages by clicks. A page that slides below the cutoff appears to vanish; open it before declaring it lost.
Averaging CTR. The CTR of a group of pages is its total clicks divided by its total impressions, never the average of the individual CTRs.
Reading a single month as a trend. One month of decline is a prompt to look. Three is a pattern.

The SEO report template shows how to put these lists into a monthly report, including how to explain a drop to a client.

Asking for both lists in Quiriz

Quiriz's Search Console connector is read-only, refreshes daily, and is included in paid plans. It reads 16 months of three datasets: [Search Console] Daily performance, [Search Console] Queries by month, and [Search Console] Pages by month (top 5,000 per month by clicks). The metrics Search Console clicks, impressions, CTR and average position are defined once, with CTR and position computed from summed totals.

The built-in example questions are close to the two lists above: Which pages got the most clicks from Google search last month? and Which queries have many impressions but a low click-through rate? Ask the comparison versions in the same plain language:

Which pages lost the most Google clicks last month compared with the month before, and what happened to their impressions and position?
Which queries with more than 10,000 impressions last month had a CTR under 2 percent and an average position better than 8?
Show queries whose clicks fell more than 30 percent against the previous month.

The answer shows the table and the filters it used, so you can check that the minimum-clicks rule and the two windows are the ones you meant. A team that wants the same threshold every month can keep it in Company Context, so the definition of a declining page does not depend on who asks.

What the data cannot tell you

No query-by-page view. Queries and pages are separate datasets, so you can see that a page lost clicks and that a query is under-clicked, but not which queries drove a particular page's loss. Open that page in Search Console to see its queries.
Anonymized queries are left out of the queries data, so query rows add up to less than the daily total.
Pages beyond the top 5,000 per month are missing, which matters for large sites.
The newest days are preliminary. Review complete months.
It shows what happened, not why. A competitor change, an algorithm update or a seasonal shift all look the same in the numbers. The reason is still your judgment.

The dashboard version of this analysis is in the Search Console dashboard post, and the next step, finding the ranking pages that earn nothing, is here.

The two lists, every month, by asking

Connect Search Console and ask which pages lost clicks and which queries are under-clicked. The answer shows its filters. Free to start.

Try Quiriz free →

Frequently asked questions

How do I find pages that lost traffic in Google Search Console?

Compare two equal windows in the Performance report, for example the last full month against the one before, on the Pages tab, and sort by the change in clicks. Then check whether impressions also fell. If they did, visibility changed; if only clicks fell, the page is shown as often but chosen less.

What is a good CTR in Search Console?

There is no single good CTR, because CTR depends heavily on position and on what else appears in the results. The useful comparison is your own: queries where your page ranks in a similar position but earns a much lower CTR than your other pages are the opportunities.

Why did one of my pages disappear from the Search Console list?

In a connected dataset that keeps only the top pages per month, a page can drop off the list because it fell below the cutoff, not because it reached zero. Check the page directly before reporting it as lost.

How do I find queries with high impressions and low clicks?

Open the Queries tab, sort by impressions, and look at the CTR column for queries ranking on page one. Queries with many impressions and a CTR well below your site norm at the same position are candidates for a better title or description.

How often should I review declining pages?

Monthly is enough for most sites, using complete calendar months. Reviewing weekly produces noise from small numbers, and a daily check mostly catches Google finalizing recent days.