By 
Shreyas M V · Published · 14 min read
Discovered, Crawled, Indexed, Ranked: How to Find Which Stage Is Broken
Name the stage before you pick a fix. Discovered - currently not indexed is a crawl problem, Crawled - currently not indexed is an index decision, and a page that is indexed can still fail to rank or to earn clicks. This guide maps each Search Console status to its likely causes and fixes, adds Next.js checks, and says where the evidence is thin.
How do you find which stage is broken?
Read the URL's status in Search Console and match it to one of five checkpoints: discovered, crawled, indexed, ranked or clicked. Each checkpoint has a different cause and a different fix, so naming it first stops you rewriting a page that has a robots.txt problem, or editing a canonical on a page Google has never visited.
- Discovered - currently not indexed means Google knows the URL but has not visited it. That is a crawl problem.
- Crawled - currently not indexed means Google visited the page and did not add it to the index. That is an index decision: check what Google can see, then duplicates and signals, then value.
- Not every unindexed URL needs a fix. Google says to expect only canonical pages to be indexed.
- An indexed page can still fail to rank or to earn clicks, and the Page indexing report cannot show that.
discovered → crawled → indexed → ranked → clicked
The rest of this page is the tree that tells you which arrow is failing. It also says which parts rest on Google's own documentation and which on practitioner opinion.
What are the five checkpoints, and where does each one show up?
The five checkpoints are discovered, crawled, indexed, ranked and clicked, and Search Console shows each one in a different place. The first three come from the Page indexing report and URL Inspection. The last two come from the Performance report.
A page is broken at the first checkpoint it fails, and no later checkpoint can be fixed until that one passes.
| Checkpoint | The question | Where to read it | What failure looks like |
|---|---|---|---|
| Discovered | Does Google know the URL exists? | URL Inspection; the Page indexing report | The URL is unknown, or listed as Discovered - currently not indexed and never visited |
| Crawled | Did Google fetch the page and see its content? | URL Inspection, Test Live URL; your server logs | The live test shows missing content, or the server answers with errors |
| Indexed | Did Google store the page in its index? | The Page indexing report; URL Inspection | Crawled - currently not indexed, a noindex tag, a duplicate, a soft 404 |
| Ranked | Does the page appear for the queries it should? | The Performance report (impressions, position) | Indexed, with no or few impressions for the queries you expect |
| Clicked | When it appears, do people choose it? | The Performance report (clicks, click-through rate) | Impressions, but few clicks |
Yoast's guide describes three of these stages as crawling, indexing and ranking, and says a page with this status stalls at indexing. The discovery checkpoint is added because Google's report separates it from crawling, and the click checkpoint because a page can pass indexing and still earn nothing.
What is the diagnostic tree?
The tree is six questions asked in order, and it ends at the checkpoint that is failing. Each question is cheaper and more yes-or-no than the one after it.
| Step | Ask | If yes | If no |
|---|---|---|---|
| 1 | Is this URL meant to be indexed? | Yes: go to question 2. | No: leave it, or mark it noindex on purpose. Feeds, pagination, filter and duplicate URLs are often expected exclusions. |
| 2 | Does URL Inspection know the URL? | Yes: go to question 3. | No: this is a discovery problem. Link to the page from a known page, list it in the sitemap, give a new page about a week, then request indexing. |
| 3 | Is the status Discovered - currently not indexed? | Crawl problem. Check server health, the size of your low-value URL inventory and overall site quality. | No: go to question 4. |
| 4 | Is the status Crawled - currently not indexed? | Index problem. Test the live URL first, then duplicates and signals, then internal links, then value. | No: go to question 5. |
| 5 | Does URL Inspection say the page is on Google? | Yes: go to question 6. | No: the Page indexing report names another reason (noindex, robots.txt, duplicate, soft 404, 404). Read the documentation for that reason. |
| 6 | Does the Performance report show impressions for the page? | Yes, with few clicks: click problem. Look at how the result reads and what else is on the results page. | No: rank problem. Compare the page with what ranks and check the query and intent. |
Kanso built this tree from three kinds of evidence: the Page indexing report documentation, the crawl budget guide, and the three practitioner pages named in the sources. The criteria were that each branch must match a status Google names, that it must be checkable with a free Search Console tool, and that cheap checks come before expensive ones.
The ordering and the fifth checkpoint are analysis, not Google guidance. Kanso did not run this tree against a set of real sites for this article, so treat it as a reading order for the documented causes rather than a tested method.
Should this URL be indexed at all?
Often it should not: Google's Page indexing report documentation says you should not expect every URL to be indexed, only the canonical pages, and that duplicate or alternate pages being excluded is usually a good sign. Settle this before you spend time on any fix.
Marie Haynes lists feeds, pagination and URLs with parameters as pages that belong in this bucket, and SEOTesting adds RSS feeds and paginated URLs such as /page/2/. Google adds that URLs blocked by robots.txt, pages with a noindex tag, duplicates and removed pages are all fine reasons for exclusion.
Two checks stop false alarms. First, search Google for the full URL, because SEOTesting reports cases where the status says excluded and the page still appears. Second, remember that the report shows at most 1,000 example URLs per status, so a list of 1,000 is a sample, not a count.
If you want a page out of the index for good, Google's removal documentation lists three ways: remove or update the content, password-protect it, or add a noindex tag. It does not list robots.txt, and Google's crawl budget guide says blocked URLs stay in the crawl queue longer. The weekly growth checklist covers how to keep an eye on the indexed count week to week.
What does Discovered - currently not indexed mean, and how do you fix it?
Discovered - currently not indexed means Google knows the URL exists and has not visited it, so the failure is in the crawl stage. A Google podcast excerpt quoted by Marie Haynes puts it simply: if Google has not visited a page, it cannot put the page in the index.
Four causes appear in the sources. A new page or site can take about a week or so to be crawled, according to the Page indexing report documentation. The other three come from Google's crawl budget guide: a slow or erroring server, a large inventory of duplicate or unimportant URLs, and low demand for the site.
Server health matters because Google raises its crawl limit when responses stay fast and steady, and lowers it when latency grows or the site returns 5xx errors or 429 rate-limit responses. Inventory matters because, without guidance, Google tries to crawl most URLs it knows about, and duplicates waste that time. Google calls inventory the factor you can influence most.
The same excerpt adds a quality angle. If Google has strong concerns about a site's overall quality, it will crawl and index fewer pages, and you will see both statuses. That means a pile of Discovered URLs on an otherwise healthy site is a crawl question, while a pile on a site with weak pages may be a quality one.
- Check response times and errors in your server logs, and fix 5xx and 429 responses first.
- Consolidate duplicate URLs, and block crawling of unimportant ones in robots.txt if you cannot consolidate them.
- Return 404 or 410 for pages you removed, and fix soft 404s. Google says a 404 is a strong signal not to crawl a URL again.
- Keep the sitemap current, and return a 304 for pages that have not changed.
- Link to the pages you want crawled from pages Google already visits.
Google says its crawl budget guide is aimed at sites with over a million unique pages that change weekly, sites with over 10,000 pages that change daily, and sites where a large share of URLs sit in this status. As of the guide's 22 July 2026 update, it adds that if your pages are crawled the day they are published, you do not need to read it.
What does Crawled - currently not indexed mean?
Crawled - currently not indexed means Google fetched the page and chose not to add it to its index, so the failure is in the index stage and the page cannot show for any query. SEOTesting stresses that it is not a penalty, and that Google does not publish the exact criteria behind it.
Marie Haynes reports a presenter at the Google Search Central event in April 2026 saying that crawling means downloading a page, and that Google may put it in a database if it thinks it useful. Per her account, the presenter gave two reasons for a crawled page being left out: Google cannot see the content, or Google looked and judged the page not good enough. The organizers asked attendees not to attribute quotes to a named Googler, so this is Haynes's account of the event and not a Google statement.
The sources disagree on how often each reason applies. Haynes writes that in almost every case she has examined the pages had quality issues, and that technical causes are rare. Yoast lists technical problems, site structure and quality side by side, and SEOTesting lists eight causes.
No source in the research quantifies how common each cause is, so the order of the checks below follows effort, not frequency. Technical checks come first because they take minutes and a wrong result costs little.
How do you fix Crawled - currently not indexed, step by step?
Work through six checks in this order, and stop when one explains the status. This is the page not indexed fix in sequence, with the source for each step.
- Confirm the page should be indexed. Skip feeds, pagination, parameter URLs and duplicates, as the section above covers.
- Test the live URL. In URL Inspection, run Test Live URL and compare what Google rendered with what a visitor sees. Haynes did this on a migrated site and found only a heading and a few boilerplate words.
- Check the signals. Look for a noindex tag, a canonical that points elsewhere, a soft 404 and a redirect, all of which Yoast, SEOTesting or Google's report name. URL Inspection shows the canonical Google selected.
- Check for duplicates. SEOTesting suggests finding which of your pages already ranks for the query in the Performance report, then using a canonical tag or merging.
- Check internal links. SEOTesting recommends linking from pages that already perform and are on the same topic, and gives site:yourdomain.com plus the target keyword as a way to find them.
- Judge the value. Compare the page with what ranks. Haynes lists four questions drawn from Google's helpful-content guidance: does it offer original information or analysis, insight beyond the obvious, more than a rewrite of its sources, and substantial value compared with the other results?
Haynes is open that an AI model cannot tell you why Google ranks a page, so treat any such check as a way to find gaps, not as a ruling. She also says a technical cause is rare, and that for most sites with many wanted pages stuck here, recovery is difficult.
| Cause | How to check | Fix | Who says so |
|---|---|---|---|
| Google cannot see the content | Test Live URL and compare it with what a visitor sees | Remove the robots.txt rule or error that hides the page, its scripts or its styles | One reported case (Marie Haynes); robots.txt, noindex and 5xx errors are also listed by Yoast |
| noindex tag or conflicting canonical | URL Inspection shows the tag and the Google-selected canonical | Remove the tag or correct the canonical | Yoast and SEOTesting; Google's report documents noindex and duplicate statuses |
| Duplicate or near-duplicate | Find which of your pages already ranks for the query | Canonical tag, redirect or merge into the stronger page | SEOTesting; Google's report says duplicates should not be indexed |
| Weak internal links (orphan page) | Count the indexed pages that link to the URL | Add contextual links from relevant pages that are already indexed | SEOTesting and Yoast |
| Low-value or commodity content | Compare the page with the pages that rank, using Google's own helpful-content questions | Add original information, experience or analysis the results lack | Marie Haynes (observation); a Google podcast excerpt quoted by Haynes; Yoast; SEOTesting |
| Search intent mismatch | Check whether the results are guides, tools or videos | Change the format to match | SEOTesting only |
| Structured data errors | Check the Enhancements reports | Fix the errors | SEOTesting only; the package holds no Google statement that ties them to this status |
| Expected exclusions | Is the URL a feed, a paginated page, a parameter URL or a private page? | Usually nothing | Marie Haynes and SEOTesting |
Two rows rest on one source each: search intent mismatch and structured data errors. Treat them as hypotheses to test, not as established causes.
Which Next.js settings can stop a page at the index stage?
Six settings in a Next.js app can stop a page at the index stage: robots.txt rules, the robots metadata field, canonical tags, parameter URLs, not-found handling and the sitemap. The table pairs each with the failure to look for and the check that settles it.
| Setting | What goes wrong | How to check |
|---|---|---|
| robots.txt (app/robots.ts) | A Disallow pattern that also matches script, style or parameter URLs can leave Google with a near-empty page. Haynes reported a site whose rule against parameter URLs blocked the files its theme needed. | Test Live URL, then compare the screenshot and HTML with a browser. A Next.js build normally serves scripts and styles from /_next/static, so check that no rule matches that path. |
| metadata.robots | A robots value set in a shared layout applies to every route below it until a child segment sets its own. | URL Inspection shows whether noindex is present. Search the layout files for robots. |
| alternates.canonical | A canonical set in a shared layout, or built from the wrong base URL, points many pages at one address, and Google then treats them as duplicates of it. | URL Inspection shows the canonical Google selected. Each page should declare its own full URL. |
| Parameter URLs | Sort, filter and tracking parameters create many URLs for one page, which Google says wastes crawling when they are duplicates. | Point their canonical at the clean URL and keep them out of the sitemap. |
| Not-found handling | A missing page that answers with a success code is a soft 404, which Google says keeps being crawled. | Request an unknown URL and read the status code. Removed pages should return 404 or 410. |
| Sitemap (app/sitemap.ts) | A sitemap that lists redirected, noindex or non-canonical URLs gives Google contradictory signals. | List only the canonical URLs you want indexed, and keep the file current. |
The inheritance rule is the one worth knowing. In the App Router, metadata is resolved by route segment from the root down, and a field that a child segment defines replaces the same field from its parent instead of merging into it. A value set once in a shared layout therefore applies to every page beneath it until a page defines its own.
This site relies on that rule on purpose: as of October 2026 its root layout marks routes noindex and the layout for /blogs overrides it. The same mechanism applied by accident is how a staging setting can stay in place in production, so grep your layouts for robots and alternates before you debug content. Each page here declares its own full URL with alternates: { canonical: PAGE_URL }.
The /_next/static path and the inheritance behaviour are how the framework works, and they are Kanso's inference applied to Haynes's case, not a finding from any source here. Check the current Next.js documentation for your version before relying on either.
What if the page is indexed but does not rank or earn clicks?
Then the problem is after indexing, and the Page indexing report cannot show it, so move to the Performance report. This is where the sources run thin: the research for this article holds little on ranking mechanics, so this section stays short and says what it cannot support.
Ranked: indexed, but not appearing
An indexed page can still be missing from your own searches. Google's Page indexing report documentation says results are customized by search history, location and other variables, so a page may not show in every search or in the same position. If Search Console says a URL is indexed and you cannot find it, Google says to assume it is indexed and eligible to appear.
Measure with impressions, not with your own search. If a page has none for the queries you expect, compare it with the pages that do appear, in the same way as the value check above. SEOTesting's intent-mismatch point applies here too: for some queries the results are guides, for others tools.
Clicked: impressions, but few clicks
When a page appears and is not chosen, the cause is in how the result reads or in what else the results page offers. Haynes notes that a results page often carries an AI answer that satisfies the searcher, which gives them less reason to click through to a page that says the same thing.
The weekly growth checklist summarises the studies of how AI summaries change click-through. Title and meta description fixes are the cheapest first test, and the Coding Agent section below covers who can prepare them.
Should you request indexing, validate the fix or wait?
Make a real change first, then request indexing or validate, and wait only if the page is new. Google's documentation says new content can take a few days to be indexed and that requesting indexing can shorten the delay. SEOTesting says requesting without improving the page rarely speeds anything up and can delay reassessment.
Validation is the second tool. After you fix every instance of an issue, you can ask Google to confirm it, and Google emails you when it has checked all the URLs and shows you its progress. Without validation the issue count still updates whenever Google recrawls an affected page.
The report also shows a Source value, either Website or Google. Google says you can generally fix only the issues whose source is Website, and that you should prioritize issues with a failed or not-started validation and a Website source.
SEOTesting says pages can stay in this status for weeks or months. Use that as a rough expectation only, because it is a practitioner statement and no source here gives a firm time limit.
What is documented and what is our analysis?
The facts below come from named sources and the interpretation is Kanso's, and keeping them apart shows how far each claim can be trusted.
- Documented (Google, as of October 2026): do not expect every URL to be indexed, only canonical pages. The report samples at most 1,000 URLs per status.
- Documented (Google): a new page can take about a week or so; requesting indexing can reduce the lag.
- Documented (Google): the crawl limit falls with slow responses, 5xx errors and 429 responses; duplicate URLs waste crawling; soft 404s keep being crawled.
- Documented (Google): the report's Source value shows whether an issue is Website or Google; indexed does not mean shown in every search.
- Reported (Marie Haynes, 16 July 2026): a presenter's account of the April 2026 event; one site whose robots.txt blocked its styles and scripts; her view that quality is the usual cause.
- Reported (SEOTesting, March 2026; Yoast, undated): the list of causes, the request-indexing advice and the weeks-or-months range.
- Analysis: the five checkpoints and the order of the six questions.
- Analysis: the claim that technical checks should come first because they are cheap, not because they are common.
- Analysis: the Next.js failure modes, which map Google's rules and one reported case onto the App Router.
- Gap: the research has no data on how often each cause occurs, no Google statement on ranking or click causes, and no test of this tree on real sites.
Where does Kanso fit into this diagnosis?
Kanso can help with finding the problem pages and preparing some of the fixes, and it cannot make Google index a page. The SEO Agent audits your site and Search Console data and turns what it finds into recommendations and tracked tasks for your approval.
The Coding Agent prepares title, meta description and canonical fixes as GitHub pull requests that you review, and it never merges or deploys. Those pull requests cover those three things, so a robots.txt rule, a server error or thin content still needs a person.
Crawled - currently not indexed: frequently asked questions
What does Crawled - currently not indexed mean in Google Search Console?
Crawled - currently not indexed means Googlebot fetched the page and Google decided not to add it to its index, so the page cannot appear in search results for any query. Google does not publish the exact criteria, and SEOTesting notes it is not a penalty. A Google podcast excerpt quoted by Marie Haynes describes it as Google having looked at the page and waiting until it is happy to look again.
Is Crawled - currently not indexed bad for SEO?
It is bad only for pages you want to rank, because a page that is not indexed gets no search traffic. For feeds, paginated pages and parameter URLs it is usually expected. Google's Page indexing report says not to expect every URL to be indexed, only the canonical pages.
How do I fix Crawled - currently not indexed pages?
Fix them in this order: confirm the page should be indexed, run Test Live URL to check Google can see the content, clear noindex and canonical conflicts, resolve duplicates, add internal links from indexed pages, then raise the page's value above the pages that already rank. Request indexing or validation only after a real change. The step-by-step section above gives the checks and the sources behind each.
Why are my pages crawled but not showing in Google?
Pages that are crawled but not showing are either not indexed, or indexed but not ranking. Check the status in URL Inspection first. If the page is indexed, Google says it can still miss your own searches because results are customized by search history and location, so read impressions in the Performance report instead.
How long can pages stay in Crawled - currently not indexed?
They can stay for weeks or months. SEOTesting says some resolve as Google reassesses site quality and internal linking, while others stay excluded when nothing about their value changes. No source in this article gives a fixed time limit.
Will these pages be indexed in the future?
Some will, but nothing in Google's documentation promises it. A Google podcast excerpt quoted by Marie Haynes says Google will take another look once it is happy, which points to changing the page or the site rather than waiting. Haynes also warns that recovery is difficult when many wanted pages are stuck.
What is the difference between Discovered and Crawled - currently not indexed?
Discovered - currently not indexed means Google knows the URL but has not visited it, while Crawled - currently not indexed means Google visited and chose not to index it. The first is a crawl problem, so look at server health, URL inventory and site quality. The second is an index decision, so look at what Google can see and whether the page adds value.
How do you handle Discovered and Crawled - currently not indexed at scale?
At scale, treat a large share of either status as a pattern to explain, not a list of pages to fix one by one. Google's crawl budget guide names a large share of Discovered URLs as a reason to read its advice, and a Google podcast excerpt quoted by Marie Haynes says to step back and consider overall quality when many pages are not indexed with no technical cause. Note that the report shows at most 1,000 example URLs per status.
How do I fix Discovered - currently not indexed?
Start with server speed and errors, because Google lowers its crawl limit when a site is slow or returns 5xx or 429 responses. Then remove duplicate and low-value URLs from what Google can find, return 404 or 410 for removed pages, and keep the sitemap current. Google says its crawl budget guide is meant mainly for very large sites, so a small site should check quality and links first.
When do 301 redirects result in Crawled - currently not indexed?
SEOTesting says redirect destination URLs can show this status because of crawl prioritization rather than a redirect error, and calls it uncommon. Its suggested fix is a temporary sitemap that contains the affected URLs. This rests on one source, so confirm in URL Inspection that the destination returns a 200 and is its own canonical before acting.
Should I request indexing or wait?
Request indexing after you have made a meaningful change, and wait if the page is new. Google says a new page can take about a week or so to be crawled, and that requesting indexing can shorten the delay. SEOTesting adds that requesting without any improvement rarely speeds inclusion.
Sources
- Page indexing report, Search Console Help
- Crawl Budget Management | Google Crawling Infrastructure, Google for Developers (last updated 22 July 2026)
- Remove a page hosted on your site from Google, Google Search Central (last updated 10 December 2025)
- Why your pages are stuck in crawled-currently not indexed. And what to do about it., Marie Haynes, 16 July 2026
- What is 'Crawled - currently not indexed' in Search Console?, Yoast
- Crawled - Currently Not Indexed: Meaning and Fixes, SEOTesting, updated 9 March 2026
Note: Facts are attributed to Google documentation or to named practitioner pages and were read on 16 October 2026. Google does not publish the exact rules for indexing, and Search Console wording changes, so confirm against the linked source before acting. Kanso has no affiliation with Google, Yoast, SEOTesting or Marie Haynes.
Search & AI discovery