Let me first start off with the title may sound more scary than it is, but I tried hard to come up with a short title for this post and I failed (traveling this week and I am tired). In short, with the new evergreen GoogleBot able to do more, some ad networks and other embeds GoogleBot can now render may render meta data, titles, canonicals, etc into your body content.
Google is aware of this and I guess is dealing with it some how.
Ryan Jones mentioned it on Twitter:
🤔 had this exact same question today! I know it *used* to be an issue, but I’m not sure how this may have changed since the evergreen WRS announcement… ping @g33konaut too?
— Kyle Blanchette (@kgblanchette) May 21, 2019
Both Martin Splitt and Gary Illyes from Google responded:
Yeah, the new evergreen WRS isn’t helping with that b/c things are kinda working as they should. Pushing stuff that belongs into the
out of it? HTML parsers should “gracefully” ignore them 😬
Now as Gary said: Who knows, maybe we can try something there…
— Martin Splitt @ 🇨🇭🏡 (@g33konaut) May 21, 2019
It’s not a large change, but rather a critical path change that may introduce canonical hijack vectors. That’s why we have this as is in the first place.
— Gary “鯨理” Illyes (@methode) May 21, 2019
So it looks like Google is onto it – and I don’t know if this is actually causing problems for Google search or not but hey, it is interesting.
Forum discussion at Twitter.