You opened Search Console, clicked into Page indexing, and found a number next to "Why pages aren't indexed" that made your stomach drop. Two hundred pages. Four hundred. On a site that only has thirty.
Take a breath. On almost every site we audit, most of those pages are supposed to be there. Several of the reasons in that list are Google's way of saying you set things up correctly. A few genuinely need your attention. Two are worth real thought.
This is the plain English translation of all fifteen reasons Google can show you, what each one actually means for your business, and, just as importantly, which ones you can stop worrying about tonight.
Jump to what you need
Read this before you read the table
A high not-indexed count is not automatically a problem. If your site has tag pages, filtered product URLs, paginated archives, old redirected links or a staging path, Google will list all of them here. A 400-page count on a 30-page site usually means Google found 370 URL variations and correctly decided not to index the duplicates.
The question is never "how many." It is "are any of my real, money-making pages in this list." That is the only thing worth acting on.
All fifteen reasons, in plain English
Google groups these into errors and normal exclusions, but that split is not very useful when you are trying to decide what to do on a Tuesday morning. This table sorts them by how much they should actually worry you.
| What Google says | What it actually means | Worry level | What to do |
|---|---|---|---|
| Alternate page with proper canonical tag | This page is a deliberate duplicate, and you told Google which version is the real one. Google listened. | Fine | Nothing. This is the system working. |
| Page with redirect | This URL forwards somewhere else, so Google indexed the destination instead. | Fine | Nothing, as long as it lands where you expect. |
| Blocked due to unauthorized request (401) | The page asked Google to log in. Usually a members area, client portal or staging site. | Fine | Nothing, if it is meant to be private. Check nothing public is in here. |
| URL blocked by robots.txt | You told crawlers to stay out of this path. | Check | Confirm the block was intentional. Accidental blocks are common after a rebuild. |
| URL marked 'noindex' | The page carries a tag telling Google not to index it. | Check | Same question: did you mean it? A stray noindex left over from staging is the classic disaster. |
| Not found (404) | The page is gone. | Check | Fine if it is genuinely retired. Redirect it if it had traffic or links. |
| Soft 404 | The page loads normally but looks empty or says "nothing found," so Google treats it as missing. | Check | Either add real content or return a proper 404. |
| Blocked due to access forbidden (403) | Your server refused Google outright. | Check | Often a firewall or security plugin blocking Googlebot by mistake. |
| URL blocked due to other 4xx issue | Some other client error Google could not categorize. | Check | Inspect one URL and see what your server returns. |
| Duplicate without user-selected canonical | Two pages look the same and you never said which one counts, so Google picked. | Check | Pick a winner yourself with a canonical tag, rather than letting Google guess. |
| Duplicate, Google chose different canonical than user | You named a canonical. Google disagreed and chose another page. | Check | Usually means the two pages are too similar. Make them genuinely different or merge them. |
| Discovered, currently not indexed | Google knows the URL exists but has not gotten around to crawling it. | Check | Normal for a few pages. A large number suggests Google is rationing attention on your site. |
| Crawled, currently not indexed | Google read the page, considered it, and chose not to include it. | Check | The most important one on this list. See the section below. |
| Redirect error | A redirect loop, a chain that runs too long, or a broken destination. | Fix | Fix it. Nobody, human or crawler, reaches the page. |
| Server error (5xx) | Your server failed or timed out when Google asked for the page. | Fix | Fix it now. If Google hit it, customers are hitting it too. |
How to read your own numbers: in Search Console open Indexing, then Pages. Scroll to "Why pages aren't indexed" and click any row to see the actual URLs. That list of URLs is the only thing that matters. Scan it for pages you actually care about.
The three that mean you did it right
These are usually the biggest numbers on the report, and they are good news.
Alternate page with proper canonical tag
Almost every site serves the same content at more than one address. A product in two categories, a printer-friendly version, a page reachable with and without a tracking parameter. A canonical tag tells Google "this one is the original." When Google follows that instruction, the copies land here.
What to look for: nothing, unless a page you want ranking is in the list. If your main service page shows up as an "alternate," your canonical tags are pointing the wrong way.
Page with redirect
Old URLs that now forward to new ones. If you have ever rebuilt a site, renamed a page, or switched from http to https, expect a pile of these. Google indexes the destination, not the doorway.
What to look for: spot-check two or three and confirm they land on the right page rather than the homepage. Redirecting everything to the homepage is a common shortcut that quietly wastes the value of those old links.
Blocked due to unauthorized request (401)
The page asked for a login. Client portals, member areas, admin screens and password-protected staging sites all belong here, and you would be alarmed if they were indexed instead.
What to look for: one thing only. Make sure no page you want the public to find has wandered behind a login.
The ones worth a look
None of these are emergencies. Each one asks you a single question, and usually the answer is "yes, that was on purpose."
URL blocked by robots.txt, and URL marked 'noindex'
These two are different mechanisms with the same question attached: did you mean to do that? Blocking in robots.txt tells crawlers not to look. A noindex tag lets them look but tells them not to list it.
What to look for: this is the single most common way a small business accidentally disappears from Google. A site is built on a staging URL with indexing switched off, then goes live, and nobody removes the block. If your traffic fell off a cliff after a redesign, check these two first, before anything else.
What not to worry about: thank-you pages, cart and checkout URLs, internal search results and login pages all belong here on purpose.
Not found (404) and Soft 404
A 404 means the page is gone and your server says so honestly. That is allowed. Pages retire, and Google handles it fine.
A soft 404 is sneakier: the page loads with a normal "success" response but has nothing useful on it, or says something like "no results found." Google decides it is effectively empty. Thin category pages with no products in them are the usual culprit.
What to look for: does the dead URL appear anywhere that still matters? If it had inbound links, ranked for something, or is still linked in your own navigation, redirect it to the closest living page. If it was a page nobody visited, let it 404 and move on.
What not to worry about: a long tail of 404s from URLs that never existed. Scrapers and bots invent addresses constantly.
Blocked due to access forbidden (403) and other 4xx issues
Your server actively refused Google. This is rarely deliberate. The common causes are a firewall or security plugin treating Googlebot as suspicious traffic, a geographic block, or rate limiting that kicks in when Google crawls quickly.
What to look for: if real pages are in this list, ask whoever manages your hosting to allow Googlebot. If you use a security plugin or a service like Cloudflare, that is the first place to check.
Duplicate without user-selected canonical, and Duplicate where Google disagreed
The first means you never told Google which version counts, so it chose for you. The second is more pointed: you did say, and Google overruled you.
Google overruling your canonical is a signal worth hearing. It normally means the two pages are so similar that Google cannot see why both should exist. The classic case is a set of near-identical location or service pages where only the town name changes.
What to look for: open both pages side by side. If you cannot explain in a sentence why a customer would need both, you have your answer. Merging two thin pages into one strong page almost always beats defending both.
Discovered, currently not indexed
Google knows the page exists but has not crawled it yet. A handful of these on a healthy site is completely normal, especially on pages published in the last week or two.
What to look for: volume and duration. If hundreds of pages sit here for months, Google is effectively deciding your site is not worth the crawling effort. That is usually a symptom of a very large number of low-value URLs, a slow server, or a site with few links pointing to it.
What not to worry about: a new page sitting here for a few days. Requesting indexing once is fine. Requesting it repeatedly does nothing.
The one everybody asks about: Crawled, currently not indexed
Google fetched the page, read it, and decided not to include it. There is no error to fix, which is exactly why this one is so frustrating.
In plain terms it usually means Google did not think the page added anything. The common causes are thin content, a page that closely resembles others on your site, an auto-generated page such as a tag archive, or a page that reads like it was written for a search engine rather than a person.
What to look for: read the page as a customer. Does it answer a real question better than the other results already ranking? If the honest answer is no, adding more words will not fix it. Merging it into a stronger page, or genuinely improving it, will.
What not to worry about: tag pages, author archives, date archives and paginated lists sitting here forever. That is normal and healthy.
The two to fix now
Unlike everything above, these affect real visitors, not just crawlers.
Server error (5xx)
Your server failed, timed out, or was too busy to answer. The reason this one is urgent has nothing to do with SEO: if Googlebot got an error, so did some of your customers. Search Console is just the alarm that happened to be loud enough to notice.
What to look for: whether it is constant or occasional. A one-off spike during a traffic surge or a host migration is worth noting, not panicking over. A steady stream means something is genuinely broken, and it is a hosting conversation, not a marketing one.
Redirect error
A redirect that loops back on itself, hops through too many stops, or points at a broken address. The page is unreachable for everyone, and the visitor usually sees a browser error rather than your site.
What to look for: these tend to appear in clusters after a migration or a URL change. Fixing the rule that created them usually clears dozens at once.
The one people always confuse with this report
"Indexed, though blocked by robots.txt"
This is not in the "Why pages aren't indexed" table. Google files it as a warning under Improve page experience, and it means close to the opposite thing: the page is in Google, but because you blocked crawling, Google had to list it without being able to read it. The result is usually a search listing with no useful description.
It also exposes a genuine misunderstanding worth clearing up: robots.txt does not keep a page out of Google. It only stops Google reading it. If you want a page kept out of search results, use a noindex tag and let Google crawl it, or put it behind a login. Blocking it in robots.txt can achieve the exact opposite of what you intended.
A ten minute triage, in order
If you only have a few minutes and want to know whether anything is actually wrong, do these five things and stop.
- Open Indexing, then Pages in Search Console. Look at the trend line for indexed pages, not the raw not-indexed number. A steady or rising line means you are fine.
- Check for Server error (5xx) and Redirect error. If either has pages in it, that is your afternoon.
- Click URL marked 'noindex' and URL blocked by robots.txt. Scan for anything you want customers to find. This is where accidental disappearances hide.
- Open Crawled, currently not indexed and look only for real service or product pages. Ignore tag, author and archive URLs entirely.
- Everything else can wait for your next review. Genuinely.
Questions we get every week
My not-indexed number is bigger than my indexed number. Is that bad?
Usually not. On sites with categories, tags, filters or a store, the number of URL variations dwarfs the number of real pages, and Google correctly declines most of them. Compare like with like: count the pages you actually wrote, and check those are indexed. That number is the one that matters.
Should I hit "Request indexing" on everything?
No. It is useful once for a genuinely new or newly fixed page. Repeatedly requesting the same URL does not increase your odds, and there is a daily limit you will burn through. If a page is not being indexed, the answer is nearly always the page itself, not the queue.
How long should indexing take for a new page?
Anywhere from a few hours to a few weeks, depending on how often Google crawls your site. Established sites that publish regularly get picked up fastest. If a page has sat in "Discovered" for over a month, treat that as a signal about the site overall rather than that one page.
The report says a page is not indexed, but I can find it in Google. Which is right?
Both, usually. The report can lag by a few days, and it reports on the exact URL, including whether it has a trailing slash, http versus https, and www versus non-www. You may be seeing a different version of the same page. Paste the exact URL into the URL Inspection tool for the current answer.
Do I need to fix every 404?
No, and trying to is a waste of a morning. Redirect a dead URL if it had traffic, had links pointing at it, or is still linked from your own site. Otherwise let it 404. A 404 is a valid, honest answer, and Google is entirely comfortable with it.
Want us to look at your report with you?
We run a free workshop every Friday at 2 PM Eastern. Bring your Search Console open and we will read your actual numbers together. No pitch, and you never have to share your screen if you would rather not.
The fifteen reasons above are the ones Google documents for the Page indexing report as of September 2026. Google renames and reorganizes these periodically, so if a label on your screen does not match one here, trust your screen and check Google's current Page indexing report documentation.