We have a guest to answer today’s webmaster video question. “In a previous video, you stated that validation of HTML wasn’t really important for ranking or SEO. Would Google really want to send its users to websites which would be broken for a lot of them?”
– Danny, how would you answer that?
– Well, a lot of websites can have invalid code but actually render just fine, because a lot of modern browsers do a good job dealing with bad code. It’s not so much that the code has to be absolutely perfect, but whether or not the page is going to render well for the user in general. It makes sense to still list these websites; it makes sense not to be that specific. And plus, people make a lot of mistakes because they’re using different HTML editors, or maybe they’re doing some hand coding. They’re little tiny things that aren’t going to have that big of an impact. If you were trying to be very very specific, I suspect, and Google would certainly know, you’d have a lot of websites that would be left out for no good reason because they would still be validating just perfectly fine.
– In valedectori. You were even joking a little bit about this.
– I did. For example, in this question, there’s a grammatical error in it, and I can understand what the question is about perfectly well. But if I were going to invalidate even taking questions based on them being perfectly grammatically correct, then it would be lost. In that case, if you will, Google can both understand and deal with pages that are perfectly grammatically correct, and perfectly HTML correct, and it can also understand when you’re being a little more casual. Because basically it understands what the page is about.
– So, Danny nailed that. I don’t really have much to add, other than usually if a page doesn’t validate, it still can look perfectly fine in the browsers, which is exactly what you said at the very beginning.
by Matt Cutts - Google's Head of Search Quality Team