When you see this, sometimes it's a temporary oddity as Google moves data around. If someone is seeing this right now for the first time, or even seeing the complete absence of the domain root on a site: query - especially when it's never been an issue before and traffic is still the same - then I'd just wait. Google is really shuffling data at the moment and in that kind of a period, stuff often happens that gets quickly straightened out.
But when you see a depressed domain root on Yahoo's site: operator as well, then I would definitely suspect a technical problem with your site. Here are some things I'd check on:
1. Does the domain root resolve directly? That is, a request for www.example.com gets a 200 response and serves content, not redirecting for content?
2. Have you addressed the common canonical issues and duplicate url issues? That means no-www redirects to the with-www version of every url and no links to the index.html version of your home pages url.
3. Do you use a "custom" 404? If so, does it return a 404 in the http server header?
4. Does every page (even the deep internal urls) have a "Home" link? If so,is it of this form? <a href="http://www.example.com/">?
|4. Does every page (even the deep internal urls) have a "Home" link? If so,is it of this form? <a href="http://www.example.com/">? |
Tedster, is there any problem without the end "/" - "slash" for the link to homepage.
any difference between http://www.example.com/ and http://www.example.com (no slash)
in my opinion, Google should not ban a site for just they do not redirect non-www to www or vice versa, because no one means to spam or duplicate their content this way. Google should consider it as one, unless they have different content.
Thanks for the info tedster.
I check over the things you mentioned and all where ok apart from:
|3. Do you use a "custom" 404? If so, does it return a 404 in the http server header? |
This was returning a 302 response to the home page, i assume wth 302 being a temp redirect its making the 404 page look like dupe content of the home page?
Anyhow have now fixed this and its returning a true 404 response.
How long is a piece of string and how long do you think it would take for yahoo ad google to remedy this hiccup which is affecting my rankings.
Would I be right to assume that if i gave google and yahoo a sitemap then this would speed up progress of fixing the errors.
It's been a long time since I saw a problem with the home page because of a missing slash. Google never intentionally banned or penalized a site for any of this - but these server configurations send out confusing signals to the algorithm. SEO involves more than on-page and off-page factors; it also involves configuring your server in a technically sound fashion. If you don't do this, and your technical oversights end up imitating a spammer profile, you can have troubles.
The sitemaps might speed the process. Glad you found that problem. Incorrect custom error pages can be like a ticking time bomb.
|Incorrect custom error pages can be like a ticking time bomb |
LOL - this one isn't ticking anymore its gone off right in my face ;)
Thanks for help!
|Does every page (even the deep internal urls) have a "Home" link? If so,is it of this form? <a href="http://www.example.com/"> |
Yikes, mine are, whats wrong with that?
Nothing is wrong with that, walrus, it's the best practice and I asked just to be sure it was checked. Some people link to index.jsp, or ../home.aspx and so on. THAT can create a duplicate url problem.
Maybe Tedster is right. I had the same problem and on Tedster's suggestion I checked and yes, in a sidebar I had a link to my home page saying, visit my home page <a href="http://www.example.com/">. After reading it here, I changed it and took out the end slash. In 2 days, the home page is the first page that shows up in a site: search.
|After reading it here, I changed it and took out the end slash. In 2 days, the home page is the first page that shows up in a site: search. |
ecmedia - I think you misread things here with regard to the end slash. Actually, the variant with the end slash is the "preferred" form, but only in the sense that if it's not there, just about all servers are set up to add it... so having it in your code saves the server a little work. I believe any change you observed due to dropping the slash was purely coincidental.
The problem is not when you link to:
It's when you link to:
That creates dupe content, which can give you trouble if people link to it and it starts to rank.
I suggest you read this thread, which is up in the Hot Topics section at the top of the Google Search forum home page....
Domain Root vs. index.html [webmasterworld.com] - another kind of duplicate
Thanks for clarifying that Tedster.