Canonicalization
-
I understand what canonicalization does, however I'm a bit confused on one point.
Generally, of course it's used to determine the main article out of two which are identical.
But what happens to the keywords if the content isn't quite identical?
Example:-
Let's say the 'first page' it is optimised for 'racing cycles'.
The 'second page' is optimised for 'second-hand racing cycles'Let's assume that the 'first page' doesn't have any reference to 'used' or 'second-hand' so it would be essentially unrelated to the 'second page'.
If I then add an canonical tag to the 'second page' that points to the 'first page' in theory, the 'second page' will drop from the search rankings and pass any link authority back to the 'first page'
What I want to know is will the 'first page', then rank for the keywords that the second page used to rank for? (in this case 'second-hand racing cycles')
-
If the contents are not identical, you don't need to worry about losing the rankings. Second-page ranking will be dropped if contents are same.
-
Hi seoman
Canonicalisation was set up by Google originally to deal with pages which were basically the same but had two different URLs so for example:
website/cycles/racing-cycles
website/cycles/productid=123If the URL contained content that was the same then you would add a canonical on the second one pointing at the first. The second one would then drop from serps and the first one would be allowed to breathe and in most cases rise because the duplicate content was taken away.
People then started to use it in a more sophisticated way and as your example shows you could canonicalse 'second-hand racing cycles to racing-cycles. This would only be in a circumstance where you believed that the content on the second-hand page was so similar to the racing-cycles page that you would find it really hard to rank for both.
So you canonicalse second-hand cycles to racing-cycles which could be a good move. The thing is that Google won't combine content from both pages it will simply rely on the content of the racing-cycles page to rank it. You must make sure that the racing-cycles page contains everything you would want both pages to be found for.
Now here's the problem.
If you canonicalse second-hand cycles to racing-cycles and the two pages are very different then Google can start to distrust your canonicals and show the page in serps anyway! (serps = search engine results pages - so they have to be very similar. It would truly be a disaster if you canonicalise one to the other and they both still ranked (badly ) but I have seen this happen.
So the rule is:
1. Only canonicalise if both pages serve the same user intent
2. Make sure that the two pages are very similar otherwise Google can ignore the canonical
3. If they are just not similar build-up the content on second-hand cycles to take it away from just racing-cycles and have it as a separate page or sub-page of racing-cycles.The conclusion is that if you want racing-cycles to rank for all the keywords and phrases that second-hand cycles does, then include them and synonyms on the page.
I hope that helps
Nigel
-
Hi seoman,
I think Nigel is spot-on here and has summarized the issues at hand well.
One thing to add: If you do deploy canonicals but are not sure how/when Google is respecting or ignoring them, the new "URL Inspector" tool in the new version of Search Console provides some helpful (and unprecedented) reporting detail on this, including URLs for "User-declared canonical" (what you set in your tag) and "Google-selected canonical" (the URL Google opted to treat as canonical).
While there doesn't seem to be any clarity as to why Google selected an alternative, sometimes the URL they picked provides a hint. We've never had this clarity from Google before on when they've opted to select a different canonical URL, so it's good to at least know when it's happening.
Best,
Mike -
Hi Mike,
That new tool is very revealing and supports my experience that you can't dupe Google into ranking a different page just by canonicalization. Thanks!
Nigel