My pages are no longer indexed by Google
A page that no longer appears in results has rarely been penalised. More often it was not crawled, or it was crawled and then judged equivalent to another. Those are opposite situations: the first needs the page made reachable, the second needs it made different.
Not crawled, or crawled then set aside
A search engine works in two stages. It crawls an address, meaning it asks the server for it and reads what comes back. Then it decides whether to add it to its index. The search console distinguishes those stages clearly, and the reason given for each excluded page is the most useful information in the whole diagnosis.
An uncrawled page is blocked by a technical obstacle: an instruction forbidding crawling, no link leading to it, a server error at the moment of the visit, or a crawl budget consumed by thousands of pointless addresses. A page crawled but not indexed raises a different question: it is too close to another page, it carries a tag saying another version is authoritative, or its content is considered too thin to deserve its own entry.
What I most often find on a shop
- Filter and sort pages generating thousands of different addresses for the same content, absorbing the entire crawl budget.
- A canonical tag pointing at another page, often generated automatically by a module and never checked.
- Product pages reusing the supplier’s description, identical across dozens of sites.
- A crawl-blocking instruction inherited from staging and never removed.
- A sitemap declaring addresses that redirect, or no longer exist.
Check page by page
-
Ask the engine about the exact address
The search console’s URL inspection tool gives the real state: crawled or not, indexed or not, and why. That is the source, not an estimate.
-
Compare with what the robot sees
The content the robot sees can differ from yours, especially if the page is built after loading by a script. A page whose text only exists after execution is harder to index.
-
Check the canonical tag
It is readable in the page source. A canonical pointing elsewhere amounts to asking not to be indexed.
-
Look at how many addresses the site exposes
If a shop with a few hundred products declares tens of thousands of crawlable addresses, the problem is not indexing: it is the multiplication of filter combinations.
Carry on with the right page
-
The canonical tag explained
What it really declares, and the most frequent mistakes.
-
The robots.txt file
What it actually controls, and what it does not.
-
The sitemap
What it must contain to be useful rather than contradictory.
-
Friendly URLs and 404 errors
If the shop’s own addresses no longer answer correctly.
Describe your need in one minute
A few targeted questions so I can reply with an estimate rather than another questionnaire.