Canonical cross domain links confusion : need some clarity on cross domain canonical tags

[mancunian]

So I have 2 websites on the same subject.

I am unsure if Google sees them as duplicates of each other. Much of the page text is very different but the page titles, description tags, layout and sitemaps are very similar.

I would like both sites to rank well.

If I do not use cross domain canonical tags on each page on site A to reference the "similar" pages on site B would both websites rankings suffer IF Google sees some of the pages on site A as similar to pages on the site B? Or would just 1 sites rankings suffer?

Obviously if Google does not see the pages as similar I would be hoping for good rankings on both sites.

If I do use cross domain canonical tags I guess one site will simply not rank at all? Possibly Google may ignore the cross domain canonical tag if it thinks the pages are not that similar?

I guess the ultimate question is how to tell if Google sees site A's content as a duplication of Site B?!

[mancunian]

I am just thinking of not using canonical tags at all as what have I got to lose? - I may get both sites with good rankings or maybe just one? (the same as if I did use canonical tags!?)

...or will Google give them both poor rankings due to possible duplication? So canonical tags at least ensure one site gets rankings?!

[not2easy]

Similar is not the same as duplicate and the canonical tag is suggesting to Google that "this" is the one you prefer to have indexed. If the page is at example.net and the similar page with the canonical tag referencing example.net is at example.com then you are telling google to index the example.net page and not the example.com page.

Canonical tags are not binding, google may ignore them. They are intended to help sort duplicate pages and not just similar/same topic pages. Some platforms like WordPress show the same (duplicate) content at several different URLs and for that, the canonical tag points to the preferred version.

The canonical tag can result in the 'non-preferred' pages not being indexed. Be careful when using signals for bots that do not benefit users/visitors. You don't want Google to think that you are trying to manipulate the value of pages.

It might help to read what Google says about how they use the canonical tag: [developers.google.com...]

[mancunian]

Thanks for the feedback.

I have read the link you mention before posting on here and I think here is the key sentence it says

"If you don't explicitly tell Google which URL is canonical, Google will make the choice for you, or might consider them both of equal weight, which might lead to unwanted behavior"

So to keeps things simple
No canonical tag
scenario 1
website A might rank at number 5 and website B might rank at number 6 = OK
scenario 2
websites A and B do not rank on page 1 as Google is "confused" as to which version to use or sees them as "similar duplicates"= BAD

With a canonical tag
scenario 1
website A ranks at number 2 as some link credit is passed on from site B which has inbound links and states website A as the canonical. Website B does no rank = OK

Other outcomes possible?
It is the "unwanted behaviour" which Google says might happen which confuses and worries me.

Any thoughts as to what Google might do are most welcome.