How to Find Pages Quietly Dying in Search Console

Content decay rarely announces itself. A page that brought 400 visits a month brings 340, then 280, then 190. No single month looks alarming, and by the time the site total moves enough to notice, you have lost a year.

The pressure on that curve increased sharply once answer engines arrived. Pew Research measured click rates falling from 15 percent to 8 percent when an AI summary is present, and with 900 million weekly ChatGPT users a growing share of research never reaches a results page at all. Decay now has two causes rather than one, and they need different responses.

Here is the method for finding it early and deciding what each page needs.

The Challenges in Spotting Decay

• Site totals hide it. A handful of growing pages mask dozens of shrinking ones.
• Seasonality looks identical. A page down 30 percent in July may simply be a page nobody needs in July.
• Percentages mislead. A page down 80 percent from 12 clicks does not matter. A page down 22 percent from 3,000 does.
• The cause is invisible by default. Losing rankings and losing clicks at the same ranking need opposite fixes.

Step 1: Pull the Right Comparison

Open Search Console, go to Search Results, and set the date range to the last three months compared against the same period a year earlier. Not the previous three months, which bakes seasonality straight into your analysis.

Switch to the Pages tab and enable clicks, impressions, click-through rate and position for both periods. Export to a spreadsheet, because the interface will not let you sort the way you need.

Step 2: Sort by Absolute Loss

Add a column for clicks lost: previous period minus current period. Sort descending.

This is the step that separates a useful audit from a wasted afternoon. Percentage change puts your least important pages at the top. Absolute loss puts the pages that are actually costing you money at the top, and that list is usually much shorter than people expect.

Work down the top twenty. That is nearly always where the recoverable value sits.

Step 3: Diagnose Each One

 

Do this before writing anything. The most common expensive mistake in content refreshes is rewriting a page whose rankings never moved, which changes nothing because ranking was never the problem.

Step 4: Decide Update, Consolidate or Retire

Update when demand still exists and the page has genuinely fallen behind. Real updates mean current figures, new sections answering questions that emerged since publication, and removing advice that has since become wrong. Not a date change.

Consolidate when several pages cover overlapping ground and compete with each other. Merge them into the strongest URL, redirect the rest, and keep the best material from each. This is usually the highest return action available and the one teams avoid because it feels like deleting work.

Retire when the topic is genuinely over. Redirect to the closest useful page. Keeping obsolete content because it once performed adds nothing and dilutes the picture of what your site is about.

The Traps That Produce False Signals

Four things reliably make a healthy page look like a dying one, and acting on any of them wastes a quarter.

Seasonality read as decay. This is why the year-on-year comparison matters. A page about tax deadlines is supposed to collapse in July. Comparing against the previous quarter instead of the previous year manufactures a crisis every single year.

URL changes splitting the history. If a page moved, Search Console treats the old and new URLs as separate rows. The old one shows a total collapse and the new one shows no history. Check your redirects before concluding anything.

Query cannibalisation inside your own site. A page can lose clicks because a newer page of yours now ranks for the same query. That is not decay, it is redistribution, and the fix is consolidation rather than a refresh of the older page.

Filtering by page instead of query. A page covering several topics can hold total clicks while losing badly on its most valuable query. Check the query breakdown for any page that matters commercially, because the aggregate hides it.

Rule out all four before you open the page to edit it. The audit is only as good as the diagnosis that precedes the work.

Step 5: Record What You Did

Keep a simple log: URL, date changed, diagnosis, action taken, clicks before. Without it you cannot tell whether your refresh programme works, and next year you will run the same audit with the same uncertainty.

Check results after eight to twelve weeks. Faster than that and you are reading noise; slower and you lose the connection between the change and the outcome.

What This Catches Early

Run quarterly, this surfaces three problems while they are still cheap. Pages sliding gently for competitive reasons, which recover well with a real refresh. Clusters of overlapping content quietly cannibalizing each other, which consolidation fixes permanently. And the newer pattern: strong positions producing steadily fewer clicks, which tells you where answer engines have moved into your category.

That third signal is the one worth watching most closely right now, because it usually spreads outward from definitional queries into the rest of a topic over several quarters. Seeing it early gives you time to move your effort toward the questions a summary cannot finish.

One habit makes all of this easier over time: run the audit on the same schedule regardless of whether traffic looks fine. Teams that only investigate during a decline are always working under pressure with incomplete history. Teams that run it quarterly build a record that makes each subsequent diagnosis faster and considerably more certain.

If you want this run across your whole library with the diagnosis attached to each page rather than a raw export, start a conversation and I will show you where your traffic is actually going.

Frequently Asked Questions (FAQs)

How often should I check for content decay?

Quarterly suits most sites. Monthly is worthwhile for large publishers where a few percent of traffic is material. Checking weekly produces noise that looks like trends and leads to unnecessary rewrites.

What comparison window works best?

Compare the last three months against the same three months a year earlier. That removes seasonality, which is the single most common cause of a false decay signal.

Should I update or delete a decaying page?

Update it if the topic still has demand and you have something genuinely better to say. Consolidate it into a stronger page if it overlaps with others. Delete and redirect only when the topic is genuinely finished.

Why do pages lose traffic without losing rankings?

Usually because an AI summary or another search feature now answers the query on the results page. Check impressions and position: if both held while clicks fell, the position is intact and the click was intercepted.

Does republishing with a new date help?

Only if the content genuinely changed. Changing the date alone is a cosmetic edit that search engines see through, and it costs you the ability to tell which updates actually worked.