403s vs 404s

[danny.wood]

Hey all,

Recently launched a new site on S3, and old pages that I haven't been able to redirect yet are showing up as 403s instead of 404s.

Is a 403 worse than a 404? They're both just basically dead-ends, right? (I have read the status code guides, yes.)

[jesse-landry]

Oh I'm sorry I clearly misunderstood the question.

I have not seen any studies or testing done on this, but I have to assume that they are ignored by spiders entirely. I certainly don't think they are more damaging than a 404 would be. A 404 tends to be ignored and only registered if a certain amount of time passes and the page is still not found. Google doesn't make it a habit to instantly remove URLs unless you ask them to.

At the very worst, the 403/404 error would de-index that particular URL but this should not affect the rankings of your other pages and your actual site. And I think it'll take at least a good 30 days before Google will stop crawling those. That said, it shouldn't be crawling them at all if there aren't any links pointing to them either internally or externally. And if there are links pointing to the pages in question, you should be redirecting them via 301. That is of course if they are links you want.

Hope this was more helpful.

[danny.wood]

Hi Jesse,

Thanks for your response!

I understand the reason the 403s are happening; I was more curious as to whether they are more damaging to rankings when hit by a spider than a 404 would be

[jesse-landry]

403s are forbiddens that are only returned if the server is told to block access to the file. If the site had been built with Wordpress in the past and has directories that match current directories, it may be returning 403 errors as the sitemap differs..

This is hard to explain and I think my wording it is confusing.

Say you had on your old site domain.com/blog/ and that went to your blog's index but now you have domain.com/blog/contents.html as your index. Well the /blog/ command would be trying to pull a directory and your server would normally automatically return a 403 forbidden for such requests.

Does this make sense? Might not be what's going on, but it's one possibility.