One of the more frustrating Search Console problems is finding that pages which had been indexed perfectly well for months, or even years, have quietly disappeared from Google. This is especially unsettling when there has been no obvious technical failure, no site migration and no dramatic change to the content. The page is still live, it loads normally, it is linked internally and it may even have been receiving organic traffic until fairly recently, yet Google no longer seems interested in keeping it in the index.
The instinctive response is usually to open URL Inspection and hit Request indexing. I understand why. It feels like the quickest way of telling Google that the page still exists and that you would quite like it back. The trouble is that a page falling out of the index is rarely fixed simply by asking Google to crawl it again. If Google has already crawled the URL and chosen not to retain it, another crawl of the same unchanged page is not really addressing the problem.
When this happens on sites we operate, I find it more useful to work backwards and ask what may have changed between the period when Google was happy to index the page and now. Sometimes the explanation is technical and relatively easy to fix. At other times, the page has become too similar to something else on the site, its search intent has blurred, or it simply no longer offers enough to justify being indexed as a separate URL.
Start by confirming that the page has actually been deindexed
A fall in traffic does not necessarily mean a page has disappeared from the index. Rankings can drop, search demand can change and Google can alter the way a result is displayed without removing the URL entirely. So before rewriting anything, I would inspect the exact page in Search Console and confirm its current status.
For one or two URLs, URL Inspection is usually enough. If you are dealing with dozens or hundreds of pages, the Page indexing report becomes much more useful because it gives you a sense of whether you are looking at isolated cases or something broader.
That distinction changes the way I would approach the problem. One article falling out of the index may genuinely be a page-level issue. Two hundred pages disappearing within the same week makes me much more suspicious of something that changed across the site, whether that was a theme update, an SEO plugin setting, a canonical rule, a migration, a server problem or a template that suddenly started outputting something Google did not like.
This is why I would resist the urge to start rewriting pages immediately. If the problem affects a large group of URLs, it is usually better to understand the pattern first. Otherwise, you can spend days improving content that was never the source of the problem.
Check the technical side before assuming the content is weak
The technical checks are not exciting, but they are worth doing because they can save an enormous amount of unnecessary work. I would make sure the URL still returns a normal 200 response, that it has not picked up a noindex directive, that robots.txt is not blocking it and that the page has not started redirecting somewhere else.
I would also look at the canonical closely. A page can have a self-referencing canonical and still not be treated by Google as the preferred version. Search Console will show both the user-declared canonical and, where available, the canonical Google has selected. If those do not match, the more interesting question is usually why Google thinks another page is a better representative of the content.
This comes up surprisingly often on publishing sites because it is very easy to create several articles around what is essentially the same search intent. An editor may see a meaningful distinction between “best indoor playgrounds in Singapore”, “best indoor play areas in Singapore” and “indoor playgrounds Singapore”, while Google may see three pages competing to answer roughly the same question.
In that situation, I would not spend much time trying to persuade Google that all three deserve to be indexed. I would compare them properly and decide whether they genuinely serve different readers. If they do, the difference needs to be obvious in the content. If they do not, one stronger article is usually more useful than several overlapping ones.
Pay attention to the exact Search Console status
Two labels that often cause confusion are Discovered – currently not indexed and Crawled – currently not indexed. They sound similar, but they point to different stages of the process.
If a page is discovered but not indexed, Google knows the URL exists but has not yet crawled it. On larger sites, I would look at the amount of low-priority material the site is generating. Tag archives, author pages, attachment URLs, filters, parameters, thin taxonomies and imported listings can all create a large number of crawlable URLs that you may not actually care about.
This is particularly relevant on WordPress sites because it is quite easy to produce far more indexable-looking URLs than the site really needs. If Google is being asked to crawl thousands of unimportant pages, I would rather clean that up than obsess over one article that has not been picked up yet.
Crawled – currently not indexed is a different situation because Google has already seen the page. If the URL used to be indexed and is now sitting in this category, I would spend much more time on the content itself, its relationship with other pages and the overall structure of the site.
This is also where repeatedly requesting indexing becomes less useful. Google has already done the part you are asking it to do. The more productive question is why the result of that crawl was different from before.
Read the page again as though you did not write it
This is the part I find most useful, especially on older publishing sites where hundreds of articles have accumulated over time.
Open the page and read it without giving it the benefit of the doubt. Ignore the fact that it once ranked. Ignore the amount of work that went into it. Ignore the keyword it was originally written around. Then ask whether the page still gives a reader something worth finding.
That does not mean every article needs original research, interviews or proprietary data. Plenty of useful pages are straightforward guides. But there should still be some reason for the page to exist beyond the fact that a keyword tool once showed search volume for the topic.
Perhaps the article contains first-hand experience, better organisation, clearer explanations, original photographs, useful comparisons or an editorial judgement that helps the reader make a decision. Perhaps it brings together information that is otherwise scattered across several sources. Perhaps it answers a question unusually well.
The pages I worry about are the ones that could disappear without losing much. They tend to repeat what is already available elsewhere, cover the topic at a surface level and offer very little that belongs specifically to the site publishing them.
A page can remain indexed for years and still eventually fall into that category. Past indexing is not a guarantee that Google will continue to treat a URL the same way indefinitely.
Look for patterns across the pages that disappeared
If several previously indexed pages have dropped out, I would almost always export them and look at them together. The pattern is often much easier to see when the URLs are sitting next to each other.
You may find that most of them belong to one old content type, or that they are short listicles, thin location pages, imported directory entries or articles built around very similar keywords. Sometimes one section of the site has gradually become repetitive without anyone noticing because every article looked reasonable when viewed individually.
It is also useful to compare the pages that dropped with similar pages that remain indexed. That can reveal quite a lot. Perhaps the surviving articles have stronger internal links, more original material, better first-hand information or a clearer purpose. Perhaps they sit closer to the main navigation. Perhaps the dropped pages all belong to a topic that is only loosely connected to what the site is actually about.
This kind of comparison is far more useful than treating every deindexed URL as a separate mystery.
Do not respond by making everything longer
A common response to indexing problems is to add more words, but I would be very cautious about that.
If an article is weak because it has very little to say, adding another thousand words of generic explanation does not make it stronger. In some cases, it makes the problem more obvious because the useful information becomes buried under a much larger amount of filler.
I would rather ask what the page is missing. Maybe the information is outdated. Maybe there is no first-hand perspective. Maybe the article needs a proper comparison rather than another paragraph of background information. Maybe the opening takes far too long to answer the question. Maybe the page is trying to satisfy three slightly different search intents and ends up doing none of them particularly well.
Sometimes improving a page means adding material. Sometimes it means cutting quite a lot. And sometimes the best edit is to stop treating it as a standalone page altogether.
Some pages are better merged than rescued
One of the hardest parts of an indexing audit is accepting that not every URL needs to come back.
If two pages cover almost the same subject, I would normally rather have one strong article than spend time trying to persuade Google to index both. This is particularly true on older sites where content has been built up over several years and different writers have approached similar queries from slightly different angles.
For example, a site might have one article about how long to spend at an indoor playground and another about how much time to plan for an indoor playground visit. There is technically a wording difference, but from a reader’s point of view the two pages may be doing almost exactly the same job.
In that case, consolidating the useful material into one page is often cleaner. The weaker URL can be redirected where appropriate, internal links can be updated and the stronger article becomes the obvious destination.
This is not the same as deleting pages simply because they are not indexed. I would not recommend indiscriminate “content pruning”, especially when it is being done because somebody has decided that fewer pages must automatically mean a higher-quality site. The editorial reasoning should come first.
Internal links can reveal how important a page really is
I also like to look at how the rest of the site treats the page.
If an article is supposedly important but has almost no internal links, sits several clicks away from anything substantial and is only discoverable through the sitemap, that tells you something about how integrated it really is with the site.
Good internal linking does not require aggressive optimisation. It usually comes from sensible editorial connections. A guide about indoor playgrounds can link to individual playground reviews. Those reviews can link to practical guides about what to bring, how long to stay, or what to expect with younger children. A broader family activity page can bring those resources together.
When those relationships make sense for readers, they also make the site easier for search engines to understand. The opposite is true as well: if nobody on the site ever has a natural reason to link to a page, I would at least ask whether that page belongs as strongly as we thought it did.
Keep the sitemap accurate, but do not expect it to rescue weak pages
Your XML sitemap should contain the canonical URLs you actually want Google to find. Redirected pages, noindex URLs and obvious duplicates generally should not be sitting there.
Beyond that, I would not spend too much time repeatedly resubmitting the same sitemap or expecting sitemap changes to force pages back into the index. A sitemap helps Google discover URLs. It does not make those URLs worth keeping.
The same applies to the Request indexing button. I use it after something meaningful has changed, not as an attempt to keep asking the same question until Search Console gives me a different answer.
If a canonical has been corrected, a noindex tag removed or an article substantially improved, then requesting indexing makes sense. If nothing has changed since the previous crawl, there is not much reason to expect a different outcome.
If hundreds of pages disappear, widen the investigation
When a large number of previously indexed pages disappear over a relatively short period, I would stop thinking in terms of individual articles and look at the site as a system.
Start with timing. When did the decline begin, and what changed around that period? Look at theme updates, plugin changes, robots.txt, canonical rules, server issues, migrations, redirect changes and sitemap behaviour. Check Manual Actions and Security Issues as well, even if you do not expect to find anything there.
Then divide the affected URLs into groups. Are they mostly older posts? Directory listings? One category? One template? One author? One particular type of article?
The more precisely you can describe the affected group, the easier the problem becomes to investigate.
If every section of the site is affected equally, I would look much harder at a technical or site-wide cause. If only one type of page is disappearing, the problem is more likely to sit within that content set.
What I would not do after an indexing drop
I would not request indexing for hundreds of unchanged pages. I would not bulk-change publication dates to make old content look fresh, and I would not add FAQ sections or several hundred words to every article simply because somebody said longer pages perform better.
I would also be very cautious about deleting large amounts of content purely in response to a coverage report. Some URLs do not need to be indexed, and a site does not become healthier simply because its percentage of indexed pages goes up.
The more useful objective is to make sure the pages that matter are easy to crawl, clearly differentiated and genuinely worth keeping in search.
That is a less satisfying metric than “100% indexed”, but it is a much better way to run a publishing site.
How long does it take for pages to return?
There is no reliable timetable, which is one reason indexing work can become so frustrating.
A straightforward technical fix can sometimes be picked up relatively quickly. A page that needs to be re-evaluated after a substantial content change may take much longer. Canonical relationships can take time to settle as well.
I prefer to track groups of URLs rather than repeatedly checking the same page. If we substantially update 20 articles, I would record what changed and then watch their indexing status, impressions, clicks, crawl dates and canonical selection over the following weeks.
That gives you a much clearer sense of whether the work is having an effect.
Checking one URL every morning usually tells you very little and makes normal delays feel like a crisis.
Sometimes the uncomfortable answer is that the page is no longer good enough
This is where the investigation eventually ends up if everything technical looks fine.
At that point, I would stop looking for a hidden technical explanation and compare the article properly with the rest of the site and with the pages already appearing for the search.
Does it say anything useful that the others do not? Is the information still current? Is the article clearly written for the site’s audience? Does the author have anything meaningful to contribute to the subject? Is there enough substance to justify keeping this as a separate page rather than merging it with something stronger?
Sometimes the answer will still be yes, and the page simply needs time to be reassessed. Sometimes a substantial rewrite is justified. Sometimes consolidation is the better option. And sometimes you discover that an article which looked perfectly respectable when it was published several years ago now feels very thin.
Publishing sites change, search results change, and editorial standards should also change with them.
So what should you actually do when previously indexed pages disappear?
I would begin with Search Console, establish whether the problem affects one URL or a whole group, rule out technical changes and look carefully at Google’s canonical selection. After that, I would compare the affected pages with similar pages that remain indexed and work out whether there is a common editorial problem.
The pages that are still useful can be improved where there is something meaningful to improve. Overlapping pages can be consolidated. Internal links can be strengthened where the connection is natural. The sitemap can be cleaned up, and indexing can be requested again once the page is genuinely different from the version Google last saw.
What I would not do is treat every deindexed page as something that has to be forced back into Google.
When pages disappear from the index, there is usually information in that change. Sometimes it exposes a technical mistake. Sometimes it shows that a cluster of articles has become too repetitive. Sometimes it points to a part of the site that has not aged particularly well.
That is why I tend to see deindexing less as a single problem to fix and more as an editorial and technical signal worth investigating. The useful work is not getting the green tick back in Search Console. It is figuring out why the tick disappeared in the first place.
Frequently asked questions
Why would Google stop indexing a page that was indexed before?
A previously indexed page can disappear because of technical changes, redirects, noindex directives, canonicalisation, server problems, duplication or changes in how Google evaluates the page. The first step should be checking the exact URL in Search Console rather than assuming the cause.
What does “Crawled – currently not indexed” mean?
It means Google has crawled the page but is not currently keeping it in the index. If the technical setup looks normal, I would review the page’s usefulness, its overlap with other content and whether it still serves a clear purpose as a standalone URL.
Should I keep requesting indexing?
Not if nothing has changed. Request indexing after correcting a technical issue or making a meaningful improvement to the page.
Should I rewrite every deindexed page?
No. Some pages need better content, some are better merged with another URL and some may not need search visibility at all. Looking for patterns across the affected group is usually more useful than rewriting everything indiscriminately.
Can internal linking help?
Yes, particularly when a page is poorly connected to the rest of the site. The aim should be to add links where they make sense editorially rather than inserting them simply for SEO.
Should I delete pages that Google no longer indexes?
Not automatically. A page can still be useful to readers even if it is not indexed. Deletion should be based on whether the page still has a purpose, not solely on its Search Console status.
How long can reindexing take?
There is no fixed recovery period. Technical corrections may be picked up relatively quickly, while content and canonical reassessment can take considerably longer. It is usually more useful to monitor a group of updated pages over time than to repeatedly inspect a single URL.
Not sure why your indexed pages are disappearing?
If you are seeing previously indexed pages drop out of Google and cannot tell whether the problem is technical, structural or content-related, The Tofu Media Group can take a look.
Our growth reviews are built around the same kind of work we do across our own publishing properties: checking Search Console patterns, indexing behaviour, internal structure, content overlap and the pages that may no longer be pulling their weight.
Request a free publisher growth review from The Tofu Media Group today!