Crawl, render, index — three steps, three failure points
People use "crawled" and "indexed" interchangeably. Google doesn't, and the distinction is where most diagnosis happens.
Crawling is Googlebot requesting your URL and receiving the HTML. Rendering is Google running the page's JavaScript in a headless browser to see what the content becomes. Indexing is Google deciding to store the result, extract its meaning, and make it eligible to appear.
Each step fails on its own. Robots.txt blocks the crawl. A client-rendered template starves the render. And a page that crawls and renders perfectly can still be refused at the index because Google judged it duplicated, thin, or not worth the storage. That last one carries no error message, which is why founders find it maddening.
Reading the Page Indexing report, status by status
This is the report to open. It lists every URL Google knows about and why it isn't indexed. Here's what each of the common statuses actually means, and the move it should trigger.
| Status | What it means | What to do this week |
|---|---|---|
| Discovered — currently not indexed | Google knows the URL exists but hasn't fetched it yet. Usually crawl capacity or low crawl demand. | Add internal links from pages Google already crawls often. Check server response times. Don't republish. |
| Crawled — currently not indexed | Google fetched the page and chose not to store it. A judgement, not a bug. | Improve or merge the page. If it duplicates another of yours, consolidate. If it's thin, make it worth storing or remove it. |
| Duplicate without user-selected canonical | Google found copies and you never declared which one is real. | Add self-referencing canonicals and fix the URL variants underneath. |
| Duplicate, Google chose different canonical than user | You declared a canonical and Google overruled it. | Make internal links and sitemap agree with your tag. Google trusts those more. |
| Excluded by 'noindex' tag | Working as instructed — or somebody left a staging directive live. | Confirm it's deliberate. This is the single most common self-inflicted traffic loss after a launch. |
| Soft 404 | The page returns 200 but reads as empty or missing to Google. | Either give it real content or return a genuine 404 status. |
| Blocked by robots.txt | The crawl never happened. | Check the file. If it's meant to be public, remove the rule — noindex won't work while it's blocked. |
The two "not indexed" statuses people confuse
These sound similar and mean opposite things, so it's worth being blunt about it.
Discovered — currently not indexed
Google hasn't looked yet. This is about access and demand: how easily Googlebot reaches the URL, how fast your server answers, and how much reason Google has to bother. Pages buried four clicks deep with no internal links sit here for weeks.
The fix is structural. Link the page from somewhere Google crawls often — a hub page, a recent post. See internal linking for building those paths deliberately rather than hoping.
Crawled — currently not indexed
Google looked and passed. Publishing more pages like it makes this worse. Common causes: the page repeats what three of your other pages say, it's a thin template variant with the city name swapped, or there's nothing on it a person couldn't get faster elsewhere.
This status across a whole page type is a verdict on the template. That's the moment to stop the content calendar and fix what's already published — the argument in when to update old content instead of writing new.
Request Indexing, sitemaps, and why new sites take weeks
URL Inspection has a Request Indexing button. It's for one genuinely urgent page — a pricing change, a corrected fact. It carries a daily quota per property, it doesn't jump any queue, and pressing it repeatedly on the same URL does nothing.
For volume, use your XML sitemap: submit it, keep lastmod honest, and let the report show the submitted-versus-indexed ratio per page type.
New domains are slow for an unglamorous reason. Google allocates crawl attention partly on demand, and a site nobody links to generates almost none. Weeks is normal; months is normal with no external links at all. No button fixes this — links from sites Google already crawls, plus a clean internal structure, are the whole answer.