Of all the rows in the Page indexing report, “Alternate page with proper canonical tag” is the one that causes the most unnecessary work. It sits in the “not indexed” table, next to genuine failures, and people reasonably conclude that a number in that table going up is bad news. It is the one status where Google explicitly tells you to do nothing.
What Google says
The page indexing report documentation defines it in two sentences, and the second one is the whole story:
This page is marked as an alternate of another page (that is, an AMP page with a desktop canonical, or a mobile version of a desktop canonical, or the desktop version of a mobile canonical). This page correctly points to the canonical page, which is indexed, so there is nothing you need to do.
Read the parts in order. Your page declares another URL as canonical. Google agreed with that declaration. The canonical page is indexed. The URL in this list is therefore not indexed, on purpose, because indexing it too would be indexing the same content twice.
The report’s own summary section says the same thing from the other direction:
Having a page marked duplicate or alternate is usually a good thing; it means that we’ve found the canonical page and indexed it.
That chart is the fastest way to keep the three apart, and getting them mixed up is the actual cause of most wasted time here. Only one of the three is Google disagreeing with you.
Why your count can be enormous
Google’s definition names AMP and mobile variants, which reads as though the status should be rare on a modern site. It is not, because the same mechanism fires for any URL that serves content and points its canonical elsewhere. On real sites the usual sources are:
- Parameter URLs. Sorting, filtering, pagination variants and tracking parameters, each serving the page and each canonicalising back to the clean URL. Google’s canonicalization documentation lists “the results of sorting and filtering functions of a category page” as a normal source of duplicates. Country storefronts with the same English copy land here too, and that is the healthy state for them; I went through the whole pattern in hreflang errors and why country versions are not indexed.
- Print or AMP versions, where they still exist.
- Syndicated or regional copies pointing home.
- Self-referencing canonicals that point somewhere else by mistake. This one is not benign, and it is the reason to read the list rather than trust the label.
A large, stable count from the first three is expected. It is not crawl waste you introduced this week and it is not a penalty. It is the deduplication system working, and the count scales with how many URL variants your platform emits, not with how good your site is.
The one case where it is not fine
Here is the failure the status name hides. Google is confirming that your canonical declaration was respected, not that it was correct. If a template bug points every page in a section at one URL, every one of those pages is “alternate page with proper canonical tag”, every one is out of the index, and Search Console will describe the whole thing with the word “proper”.
I have seen this twice, and both times it came from the same place: a canonical tag hardcoded into a template instead of being generated per URL. Nothing errors. Nothing is flagged. Traffic to that section simply stops, and the report says everything is working as intended, because from Google’s point of view it is.
So the check is not “is the number big”. It is “are any pages I wanted indexed sitting in this list”.
Finding the ones that are not fine
Five passes over the list, and you are looking for exceptions, not totals.
- Open the Page indexing report, click into “Alternate page with proper canonical tag”, and read the sample URLs rather than the count. Google caps that sample at 1,000 and does not guarantee it shows every URL in a status.
- Scan for anything without a parameter or a variant pattern. Clean, ordinary URLs in this list are the ones to question. A URL with
?sort=pricebelongs here;/services/technical-seo/does not. - Inspect one of the suspicious ones. Compare User-declared canonical against Google-selected canonical in the URL Inspection tool. If they match and the target is wrong, the bug is in your template, not in Google. If they do not match, the page belongs to a different status entirely, duplicate, Google chose different canonical than user, which records a disagreement rather than an agreement.
- Open the page and view source. Count the canonical tags. More than one, or one pointing at a page with different content, and you have found it.
- Confirm the canonical target is actually indexed. The whole status rests on that assumption, and if the target is not indexed then nothing in this cluster is in Google at all. That case is a different problem, and it starts at URL is not on Google.
What to do with the number
Nothing, in almost every case. Do not add noindex to these URLs, which would be telling Google to drop pages it is already handling correctly, and can muddy the canonical signal you meant to send. Do not block them in robots.txt either, for the reason I keep repeating in noindex vs robots.txt: a page Google cannot fetch is a page whose canonical tag Google cannot read.
If the volume genuinely bothers you on a large site, the fix is upstream. Emit fewer URL variants, keep parameters out of internal links, and the count falls on its own. That is a platform change, not a Search Console change.
The broader habit is the one I set out in the guide to every indexing status and applied again this week to page with redirect: the “not indexed” table is a description, not a scorecard. Three of its largest rows exist to tell you that deduplication worked. Read the URLs, not the totals, and spend your afternoon on the statuses where the pages under them are pages you meant to rank.
Shopify stores see this status in bulk, because collections give every product a second working URL that correctly declares the first. That is the system working, and the Shopify collections version explains which of the two canonical statuses is the one actually worth reading there.
Sources
- Google Search Console Help, page indexing report, definitions of “alternate page with proper canonical tag”, “duplicate without user-selected canonical” and “duplicate, Google chose different canonical than user”, checked 28 August 2026
- Google Search Central, what is canonicalization, on duplicate clustering and sorting and filtering variants
- Google Search Console Help, URL Inspection tool, on comparing the user-declared and Google-selected canonical