Crawled, currently not indexed means Googlebot fetched your page, read whatever the server returned, and decided not to add it to the index. Unlike Discovered, this is a decision about the page itself. Google has a copy and does not want it. The status also says the page may or may not be indexed in the future, which is Google's way of saying it will look again if something changes. Something has to change. Requesting indexing on the same page again asks Google to make the same decision again, and it will. This article covers the four causes I see on AI-built sites, the four-step check that tells you which one you have, the rewrite that gets a page in, and why Validate Fix keeps failing.
First, read what Google actually read
Before you diagnose anything, look at the copy Google kept. Open Google Search Console, paste the URL into URL Inspection, and click View Crawled Page on the result. Three tabs appear. HTML shows the source Google stored. Screenshot shows how Google rendered it. More Info lists the resources Google could not load. Read the HTML tab as a stranger. If your headline and paragraphs are in there, Google judged your content. If the body is an empty div, Google judged an empty page, and no rewrite will help until the rendering is fixed.
The crawl date on the same screen matters too. If Google last crawled the page before your most recent edit, the decision was made about the old version. Update the lastmod date in your sitemap, request indexing once, and wait for a new crawl before you judge the content.
The four causes on AI-built sites
I have read a lot of these reports for sites built with Lovable, Bolt, Cursor, v0, and Replit. The cause is one of these four almost every time, and the crawled copy tells you which.
1. Google read an empty page
On an older client-side build, the crawler fetched the HTML shell and, in the first pass, saw a title and an empty div. It may have rendered the JavaScript later, or it may not. Either way, what it kept a record of was a page with no words. If your content is not in the HTML tab of the crawled copy, Google never saw it. Fix the rendering first; the Lovable stack question covers how to tell and what to do. On that build the fix is an upgrade, not a rewrite.
2. The page is a near copy of another page
Ten city pages that share one paragraph with the city name swapped. A services page that restates the homepage. A blog post the builder generated from the same prompt as three others. Google crawls them, notices the overlap, keeps one, and files the rest under this status. It is not a duplicate status, because the pages are not identical, but the decision is the same: the index already has this. The check is simple. Put two of the pages side by side and count the sentences that differ. If most of the page is shared, Google is right, and the fix is to merge them or to make each one say something the other does not.
3. The page does not answer anything
AI builders write copy that sounds professional and says nothing. Welcome to our site. We are passionate about quality. Contact us today. Google has more pages like that than it can use. A page earns its place in the index by being the best available answer to some question. Before you rewrite, open the page and ask what question a stranger would have typed to deserve landing here. If you cannot name one, the page has no reason to exist in the index, and Google agreed.
4. The page is fine and the site is not
A young domain with no links from anywhere gets a small allowance of indexed pages. Google fills that allowance with the pages it trusts most and leaves the rest here even when they are decent. This is the only version where the fix is outside the page: earn a few real mentions, link the page from your homepage, and give it time. You can recognise it because the page passes the other three checks and the site as a whole has a long list of Discovered and Crawled URLs together.
How to tell which one you have
- Open the crawled HTML. If your text is missing, it is cause one. Stop and fix rendering.
- If the HTML is complete, open the page next to its closest sibling. If the two share most of their sentences, it is cause two. Merge them or rewrite one.
- If the page is unique, read it as a stranger and write down the single question it answers in the first two lines. If you cannot, it is cause three.
- If the page passes all three, it is cause four. Add a homepage link, request indexing once, and check again in three weeks.
A quick way to check cause two across a whole site is to copy the first paragraph of each page into a text file and search for repeated sentences. On a site with location pages, the repeated sentence count is usually most of the page.
Want a second pair of eyes on the page?
Send me the URL that is stuck in Crawled, currently not indexed. I will read the crawled HTML and tell you which of the four it is, with the rewrite direction if it is the content.
The rewrite that gets a page in
The rewrites that move a page out of this status share a shape, and it is not length. The question the page answers appears in the H1 and is answered directly in the first paragraph. The rest of the page adds what a competent person would add: specifics, numbers, a process, the exceptions, the case where the general advice is wrong. The page links to two or three related pages with real anchor text, and two or three pages link back. And it stops sharing its paragraphs with any other URL on the site.
Work through it in this order for one page before you touch the rest.
- Write the H1 as the question or the exact phrase people type. Check Search Console's Queries tab for the words, not your own vocabulary.
- Answer it in the first two sentences. No introduction, no history of the company, no welcome.
- Add three things a generic page would not have: a number you can stand behind, a step-by-step, and an exception or a case where the answer changes.
- Delete every sentence that could appear on a competitor's site unchanged. If the sentence would be true of any business in your category, it is not helping this page.
- Add two internal links out to related pages and edit two related pages to link back, in the body text, with descriptive anchors.
- Update lastmod in the sitemap, request indexing once, and leave it for three weeks.
On one of my own review sites, a review published on 3 August earned 1,359 impressions in Google's AI features within eleven days, on a domain under six months old. Four of the eight most-cited pages on that site are literal question titles that answer themselves in the first two sentences. The pages did not get in because they were long. Ahrefs found a correlation of 0.04 between word count and citation position, and that more than half of cited pages were under 1,000 words. They got in because each one answers one question that no sibling page answers.
What Crawled, currently not indexed is not
It is not a penalty; those appear under Manual Actions. It is not a crawl budget problem; Google spent budget and fetched the page. It is not a speed problem, and Core Web Vitals do not decide indexing. It is not a missing schema problem; structured data describes a page Google already wants, it does not make Google want one. And it is usually not a settings problem, which is why the settings tools in your builder cannot fix it. The decision is about whether the page adds anything the index does not have. That is the whole test, and it is the same test as getting cited in Google's AI Overviews, which is why I do the two jobs at once.
Questions people also ask
How long does Crawled, currently not indexed last?
Until the page changes or the site's authority changes. Google does recrawl these pages, typically within a few weeks on an active site, and re-decides each time. A page that stays the same stays out.
Should I delete pages in this status?
If the page is a near copy or answers nothing, yes, or merge it into the page that does the job. Fewer, better pages raise the priority of everything left. If the page is unique and useful, keep it and improve it.
Does Crawled, currently not indexed hurt the rest of my site?
Not directly. A large number of them is a symptom that the site has more pages than it has distinct things to say, and that does lower Google's appetite for crawling the rest. Cutting the weak pages helps the strong ones.
Why is my new blog post Crawled, currently not indexed?
Most often because the post reads like the other posts on the site, or like the thousand other posts on the same topic. Check whether the first two sentences answer the question in the title. Then check whether anything in the post could only have been written by you.
Does requesting indexing again help?
Only after the page has changed. Google will re-read it and re-decide. Requesting indexing on an unchanged page asks for the same verdict and gets it.
If the page is in the other queue, the one where Google has not fetched it at all, the fix is different and mostly about links. That is Discovered, currently not indexed. And if the page is indexed but nobody sees it, the problem has moved from indexing to ranking, which is Why is my website indexed but not showing on Google?.
