Diagnosing a traffic drop
A sudden fall is often explained by deindexing rather than by lost positions.
What is indexing? Definition and how it works
Indexing is the step where a search engine records a page in its database after discovering and analysing it. A page that is not indexed cannot appear in any result, however good its content: as far as the engine is concerned, it does not exist.
In one sentence
Indexing is the moment an engine decides to keep your page in its database — without it, no ranking is possible.
Key points
- Discovery, crawling and indexing are three distinct steps: a page can be seen without being indexed.
- Indexing is the engine's decision, not a right: it can refuse a page it judges to add nothing.
- A noindex tag left behind after a redesign is enough to erase a site from results.
- Search Console states exactly which pages are indexed and why the others are not.
Term at a glance
- Indexing
- Indexation
- French term
- Indexation
- Domain
- Technical SEO
- Category
- Search
- Level
- Beginner to intermediate
What exactly does "indexing" mean?
A search engine works in three stages. It discovers that an address exists, through a link or a sitemap. It crawls it, meaning it downloads and interprets the page. Then it decides whether to index it. Those stages are independent: being crawled does not guarantee being indexed.
That decision is not automatic. The engine judges whether the page adds something: original content, usefulness for a query, no duplication of a page it already holds. A page repeating another's content, or with almost nothing to say, can be crawled and then set aside.
This is why the Search Console "Pages" report separates several statuses. "Discovered, currently not indexed" means the engine knows the address but has not judged it worth crawling. "Crawled, currently not indexed" means it read the page and chose not to keep it.
How does indexing happen?
- 01
Discovery
The engine learns the address exists through an internal link, an external link or your sitemap.
- 02
Crawling
It downloads the page, may execute the JavaScript, and interprets the resulting content.
- 03
Analysis and judgement
It assesses originality, checks directives (noindex, canonical) and looks for duplicates.
- 04
Indexing or rejection
If it keeps the page, it joins the database and becomes eligible to rank. Otherwise it stays invisible.
A concrete example
A company rebuilds its site and forgets to remove the noindex tag from the staging environment. The pages are live, fast and well written, but each tells the engine not to keep it. Three months later organic traffic has collapsed and nobody understands why: the content was faultless, one line of code had made it invisible.
Why watch your indexing
Validating a rebuild
After a migration, checking that the new addresses are indexed is the first control to run.
Spotting duplicates
Pages excluded as duplicate content reveal a structure that needs clarifying.
Saving crawl budget
On a large site, stopping the engine spending time on worthless pages benefits the important ones.
What you control, and what you do not
- You can make discovery easier: sitemap, internal linking, external links
- You can clear the obstacles: directives, redirects, server-side rendering
- You can request reindexing after a fix
- You cannot force indexing: it is the engine's decision
- The delay varies, from hours to several weeks
- A page can be deindexed later if it loses relevance
Why indexing matters to a business
It is the prerequisite for everything else. Investing in content, link building or performance has no effect on a page the engine did not keep. Before trying to rank better, check that you are in the race at all — and that check takes minutes in Search Console.
Frequently asked
How long does indexing take?
From hours to several weeks. An established, frequently updated site is crawled more often than a new or rarely changed one.
How do I know whether my page is indexed?
Search Console gives the reliable answer, page by page, with the exclusion reason where applicable. Searching the exact address gives a quick but less precise indication.
Should every page be indexed?
No. Thank-you pages, internal search results, print versions and private areas are better excluded. Focusing crawling on what matters improves how useful pages are handled.
Related terms
Sources and references
Not sure how many of your pages are actually indexed? That is the first diagnostic to run before any SEO work.
Request an indexing audit