For about two months, business owners and SEO professionals have reported that pages have been removed from Google’s index without a clear explanation.
These reports began in late April and continued into June. In many of those reports, the affected pages had no manual action or error code. They moved into the “released” or “hidden, not yet identified” buckets and stayed there.
Google said it saw nothing unusual in the data. A more detailed independent investigation comes from Glenn Gabe, who tracked the full decline of one site in the index.
Many of these reports do not reduce at all. They measure losses, legal options, or reporting noise filed under the same name.
If you learn your mistakes and act on them, recoverable losses can be permanent losses.
What SEO Benefits Are Reported
The current wave follows a late April question from Pedro Dias, a former Google employee. He asked if others were seeing pages leaving the index at a higher rate. Many say they are, describing the same pattern.
The condition that does most of the work in these reports is “crawled, not yet crawled.” It means that Google downloaded the page and chose not to index it. That is different from the page that Google has found but is not yet clear.
Some accounts have described all the properties that go to that state instead of just a few URLs. One site owner reported that almost every site was removed after the basic March update. Another, listed for six years, watched each page turn to the same shape.
John Mueller from Google presented the reports this week. He described the move as normal and said he did not see anything out of the ordinary. The domain owners did not get that confirmation, because the reports were coming from many places at the same time.
When Reports Are Equal
Google’s 2026 ranking calendar was packed. Spam review and main review started in March, and comprehensive review started in May. We covered how the May update reset visibility, with Reddit gaining top positions in every single niche the marketer tracked.
Two months earlier, Amsive received a March update removing visibility away from aggregators. The same types of sites were moved in opposite directions across the two reviews.
Major updates change rankings, and ranking changes are easy to mistake for deindexing. A page that loses impressions still sits in the index. None of this proves that the updates caused the reports, but it does explain the noisy background they came up against.
This is not the first time that Google has attributed the removal of large amounts as a quality or perception problem. Earlier, Gary Illyes said that a high number of “crawling, currently unindexed” URLs “could indicate general quality issues,” and explained the circumstances in which Google’s view of a site has changed. That is an example, not a description of this year’s reports.
What to do
First, Verify that the Data is True
Before you classify anything, make sure the data is real. Search Console has had some reporting issues this year.
Google’s Data Anomalies page lists a logging error that misreported impressions from May 2025 until the end of April 2026. The fix applies going forward, and Google said it will not restore historical data.
The error of perception increased the statistics, so the correction is seen as a decrease. A site that saw impressions drop in early May may be learning to adjust instead of losing visibility. Clicks are not affected by that error, making click data your strongest signal in this window.
One clean test compares the previous bug window to the debug window posted in the Performance report. Focus your click practice on GA4 live sessions to see if real traffic is driving. Reported anomalies are located in the Performance and Discover reports. Page Indexing report and URL test are not included among them.
To verify that a particular URL is actually indexing, Google’s URL checking tool is a well-documented method. “Site:” search is a condition check, not a reliable reading of index condition.
The Difference That Determines Everything
Most of the diagnostic work is to sort a single symptom into the correct cause. “My pages are gone” can mean several different things, and the answer varies for each.
True deindexing means that the URL that was indexed now does not exist. You confirm it in the URL Test, where the condition reads not specified and gives the reason. This is what descriptive reports are, and it’s worth checking before you think about it.
Loss of quality is the same look. The page is always indexed but shows up with less or fewer questions. After the main review, that’s the most common result. The page is still there. It gains few impressions, which can be displayed by the dashboard as a cliff. We’ve covered why the “found, currently not displayed” status can persist for reasons unrelated to the penalty.
Canonical integration is the third issue. Google stores the content but gives a different URL, so the page you selected reads as unselected. In checking the URL, that appears as a duplicate where Google has chosen a different canonical than the one you set. The technical blockade is the fourth. A missing noindex, a robot rule, or a server error can pull a page without any algorithmic judgment behind it. Martin Splitt walked through how a page goes from being found to being indexed, and many “missing” pages fail at a step you can name.
The fifth case is a reporting artifact, and the correction of views above is a live example. Gabe’s investigation is a useful model. He worked on his Search Console sites until the root cause was discovered. A hand gesture that was not seen at first appeared later. In that case, early absences did not rule out another. Confirm the situation, find the failed step, and execute.
Why This Is Important to Your Research
Reports cover specific site types, so your exposure depends on what you do.
Publishers and web sites have very large footprints, and small or template pages are the first to look usable. If you’re running thousands of similar pages, sample from URL Inspection rather than relying on an aggregated count. The count can go for reasons unrelated to quality.
Ecommerce sites often have unique and different URLs that overlap. Pages may appear as unselected rather than deleted, so confirm before treating them as lost. Affiliate and comparison sites are close to the quality line, where “obvious, yet unspecified” issues often come together.
Local and service area sites address this with their site pages. A set of nearly duplicate landing pages is the type of template that Google tends to skip first.
If your index count falls there, sample a few of those URLs in the URL Test before you react. Fixing small site pages is to consolidate or strengthen them, not to enter a panic ticket.
Agencies have a more difficult version of the job. A panicked client says, “We’re not on Google,” when the truth is often less. The first move is to confirm the scope, then confirm the cause. A site that has lost ten percent of a small portion is one conversation. The site that lost money pages one by one.
The most dangerous cases at this time are the parties who act before they confirm. Others may add noindex to “reset” pages, reorder URL paths, or fill out emergency tickets. It all depends on the chart which may be a reporting artifact. Each of those moves can make a temporary problem permanent.
There is no trick to restore pages. Google and SEO experts keep pointing to strong page value, clear canonical signals, and clean crawling methods. None of this is guaranteed, and nothing is imminent.
Bottom line, nothing helps unless you fix the problem you actually have.
What we don’t know
The cause is not confirmed. Google did not announce a change in indexing behavior, and Mueller described the move as normal.
Treat any single explanation, including the theory of AI content running through SEO platforms, as a hypothesis rather than a finding. Nothing in Google’s public comments linked those reports to AI findings. Time also overlaps with major updates that move the levels themselves.
There is also no reliable public measure of the level of truth. Public reports show direction, not size, and the latest news reports add noise to anyone trying to size it up. Even a large number of public reports do not meet the estimated value.
Looking Forward
A confirmed update will resolve this. So is Google’s statement about index selection, or the clean expansion of reporting data.
If the index entry bar actually rises, the divergence becomes sharper. Sites with unique content are static, and sites that use large volumes of similar pages are not. That’s still an idea you can test against your pages, not a finding.
Until Google confirms the cause, the status is available before the action. View your URL test results on a sample of affected pages. Keep click data as your anchor while impression reporting is stable. Treat the index number as a number to verify rather than one to trust.
Additional resources:
Featured Image: GaudiLab/Shutterstock
