An International SEO Conundrum
-
Hello all,
I'm looking for opinions on this.
Imagine there is a website example.com in English and the company 'Example' wanted to translate some of the pages (not not all) in to Russian.
So they set up example.com/ru and translate the key pages into Russian. But half of the pages on.com/ru are left in English and there are no plans to translate them.
How would you handle the pages in Engish on .com/ru?
My thoughts are that they should:
- Canonicalise to the same versions on .com, and...
- Remove RU hreflang tags from the pages on .com/ru which are in English
Otherwise, users searching in English with Russian browser settings could land on a page in English but then navigate to a translated page in Russian (+the menu navigation items will be in Russian) = bad UX. Not to mention they would be telling Google a page is in Russain but Google would be crawling English.
So IMO, the best option is to use canonicals for this so that the .com version of the page is indexed. Then when a user lands after searching in English they will always be served English pages within that session.
If English speakers/searchers land on the .com/ru page that would lead to a website half in one lang and half in another.
I'm aware that Google recommends not using rel="canonical" across country or language versions of your site, but I believe they are making that recommendation based on an assumption that all pages are going to be translated to another language.
In this case, there is no intention to do that, ever.
Thanks for your thoughts and opinions.
Cheers, Gill.
-
I would do exactly the same... but I would do more: if a page in the /ru/ subfolder is not translated to Russian, then I would see if it's possible to not even create a clone that I must, then, canonicalize.
I see "I would see" because I don't the website, hence I don't know if there's a business reason behind this useless duplication.
Said that, my suggestion is the best one, sincerely, also for avoiding crawl budget waste (rel="canonical" doesn't prevent a page from being crawled).
Regarding you worry, you should not worry about it because this is a plausible case for using the rel="canonical".
However, if you're fearing it, you have also this option: implementing the hreflang and setting it up to "en-ru" for the English pages targeting Russia and simply "en" in the ones targeting the English speaking users in the rest of the world.
In that case, you should avoid the cross geo-targeted pages canonical, and implement the rel="canonical" as if you were working on two separate websites.
-
Thanks for the advice Gianluca!
That would be ideal (not even create a clone). The problem with that approach is that the UX would break if a user lands on the Russian versions and then tried to navigate to a page which doesn't exist (like a none translated page in the top nav). How would you handle that?
Not too worried about crawl budget as it's a very small site. Would you agree there is no cause for concern for a site which only has around 120 pages?
Good idea re "implementing the hreflang and setting it up to "en-ru" for the English pages targeting Russia and simply "en" in the ones targeting the English speaking users in the rest of the world."