Technical foundations
Fixing pages Google will not index, one reason at a time.
A page that is not in Google's index cannot rank for anything. Search Console tells you which pages are out and gives a reason for each. The reasons are different problems with different fixes, and treating them as one is how sites make it worse.
Where the answer lives
Search Console's Page indexing report sorts every address Google knows about on your site into indexed and not indexed, and gives a reason for each page that is out. That list of reasons is the diagnosis. Two sites with the same number of excluded pages can have completely different problems, one a technical switch left in the wrong position, the other a site full of pages Google does not think are worth showing.
Google limits the example addresses it shows for each reason to 1,000, and says the list is not guaranteed to be complete even below that. So the report is read as a sample of each pattern, and the pattern is confirmed with a full crawl of the site.
Signs you have an indexing problem
A service page that never appears
Searching your exact service and town shows competitors, directories, even your old blog posts, but never the page built to win it. Check whether Google has that page at all.
The indexed count falls while you publish
More pages going live each month while the number Google shows stays flat or drops means new work is going into a site Google is reading less of.
Search Console shows thousands of addresses you never made
Filters, tags, calendar pages, printable versions, tracking parameters. The site is generating addresses faster than anyone is writing pages.
Google shows the wrong version
The http version, a version with a trailing parameter, or an old address instead of the new one. Google has chosen a different canonical from the one you intended.
The reasons, in Google's words, and what each needs
Crawled, currently not indexed
Google says the page was crawled but not indexed, may or may not be indexed in the future, and there is no need to resubmit it. The fix is the page itself: merge overlaps, improve thin pages, link to it properly, or keep it out on purpose.
Discovered, currently not indexed
The page was found but not crawled yet; Google typically wanted to crawl it but expected that to overload the site and rescheduled. Often a sign of a site with far more addresses than it needs, or a slow server. Reducing junk addresses and improving internal links usually helps.
Duplicate without user-selected canonical
Google considers the page a duplicate, you did not say which version you prefer, so Google chose the other page. Fix it by marking the preferred page with a canonical tag, or by merging and redirecting.
Excluded by noindex tag
Google found a noindex instruction. If that was deliberate, nothing to do. If it was not, which is common after a redesign or a plugin setting, remove it and the page can be indexed.
Soft 404
The page tells visitors it is not found or empty, but tells Google everything is fine. Either make the page genuinely useful or return a real not-found response.
Page with redirect and Not found (404)
Usually correct: old addresses that redirect or no longer exist. Worth checking only when an address that should be live shows up here, or when important old pages return 404 instead of redirecting to their replacement.
A brokerage buried by generated pages
A Las Vegas property brokerage's listings software had generated tens of thousands of pages: sold homes, expired listings and every filter combination. Google was spending its time on those, and 237 pages of real content were being buried. Every page got a written decision, the listings vendor's part was packaged and chased until it was done, and the real content was given a clear structure.
By 2 September 2026, Ahrefs counted 27 of its pages quoted in AI answers, other sites linking in had grown to 570 from 66 in March, and visits from Google in the 28 days to 31 August were 778, from 497 the month the work began.
A law firm's refused pages, worked back down
A Las Vegas injury firm lost thousands of pages from the index within weeks when Google changed how it treats older sites. 4,144 were refused. Overlapping posts were merged into single strong pages, spam links were disowned, and every remaining page got a reversible decision. By 3 September 2026 the count was 2,549, and not one page had been bulk deleted.
How I work through an indexing problem
Separate penalty from judgement
First, the Manual Actions report. If it is empty, nothing is being punished, and the work is about the pages, not an appeal.
Group by reason, then by folder
Each reason is exported and grouped by the part of the site it comes from. Patterns appear fast: a tag archive, a listings folder, an old blog, a set of templated town pages.
Inspect examples from each group
A few pages from each group go through the URL Inspection tool to see when Google last crawled them, which page it chose as the main version, and what it actually saw.
Decide per group, in writing
Improve, merge and redirect, noindex on purpose, or leave. Dated and reversible, so next month can prove whether each decision worked.
Fix the causes, not only the pages
If a plugin keeps creating thin archives or a feed keeps generating filter pages, the setting that makes them is fixed too, or the problem refills.
Validate where Google allows it
For statuses with a fix to validate, Search Console's Validate fix is used; Google says validation typically takes up to about two weeks but can take much longer.
Indexing problems after a redesign or move
The most sudden indexing problems follow a redesign, a platform move or a domain change. A noindex setting left switched on from the staging site. Canonical tags pointing at the old domain. Redirects missing for pages that ranked. Internal links still pointing at old addresses. Each can take important pages out of Google within days.
These are also the fastest to fix, because the cause is a setting rather than the quality of the pages. The first check after any launch is the Page indexing report and the Manual Actions report, then a crawl of the live site looking for exactly these four things.
Why bulk deleting is the wrong reflex
When a report shows thousands of excluded pages, deleting them looks like tidying. It throws away links other sites gave those pages, searches they ranked for that nobody noticed, and history Google had with them. It also makes the number fall without fixing why the pages were refused, so the next batch of pages follows them out.
Every indexing problem I have fixed came down by decisions made page group by page group. The count falling is the record of the work, not the goal of it.
What a healthy index looks like
Not every page belongs in Google. Login pages, thank-you pages, internal search results and thin archives are better left out. A healthy site has its services, towns, guides and useful answers indexed, and the rest kept out deliberately, by noindex or by not creating the addresses at all. The goal is not the biggest possible number of indexed pages. It is the right pages in and everything else out on purpose.
That is also the test I apply at the end of the work: the pages that earn are in, the pages that do not are out by decision, and the report shows fewer surprises each month.
Questions about pages not being indexed
Is a falling index count a penalty
Rarely. A manual penalty appears in the Manual Actions report and names the problem. Most falling counts are Google's judgement about pages, which can be changed.
Should I request indexing for every page
No. For pages Google has crawled and chosen not to index, its own help says there is no need to resubmit. Request indexing for pages you have genuinely changed.
Will a sitemap fix indexing
A sitemap tells Google a page exists. It does not change Google's verdict on a page it has already read.
How long does recovery take
Weeks to months, depending on how many pages are involved and how often Google recrawls the site. The work is recorded monthly so progress can be seen as it happens.
Can you fix it if my platform generates the pages
Yes. Some fixes are settings on your side; others belong to the platform or vendor and are written up precisely and chased until they are done.
Is Google ignoring your pages?
Send me the site. I will read the indexing report and tell you which reasons you have, and which ones actually matter.