|Changing website domain name and resulting problems on Google|
Google Impact of 301 redirects & sand box issues due to domain name change
We had a site up for about 3 years with about 8000 items. Home page had a PR ranking of 4 and most pages had PR 3 inside. The name of the site was very restrictive and we kept adding product that was way outside the name. We felt we had to change the name to something that could sell anything.
We launched the site with the new domain name about 4 weeks ago. The new site has about 40000 items. We made a 301 redirect for each of old pages to correspond to each of the new pages on the new site. After a few days, all of the references to the old site on Google disappeared.
We have read a lot about the problems people are having with new domains and 301 redirects since we saw this problem...I wish we had never done it.
But, here is my question. Our site is being spidered everyday by Google...about 1500 hits daily by GoogleBot. Per the webmaster tools, the home page gets reviewed regulary. When we search in Google using the following....... site:www.mysite.com, we get 1160 results. But, if you search in google just using mysite, you only get the home page and the other websites where our adword advertisement has shown up.
Why can we not reach our site if site:www.mysite.com reveals 1160 items? And if you have any other insights as to how long this process of getting stuff up in Google will take or other relevent info, please let me know.
A very concerned internet website owner.
Hello golfnfuul, and welcome to the forums.
I assume you've read by now that when you changed the domain name, you essentially began a new website -- in Google's eyes if not in your own. You were right to use the 301 redirects, of course. They will help. But still, what you are experiencing is the kind of thing people have called the "sandbox effect" - the interaction of various filters and trust measures that slow down the appearance of new domains in the regular search results.
Today it is possible to get beyond some of this a bit faster than it was, say, a year ago. The most relevant thread I can point you to is this:
Filters exist - the Sandbox doesn't. How to build Trust. [webmasterworld.com]
Understanding factors that restore and maintain results
I read the sandbox thread you referred to a few minutes after posting. I think that it is fair to say, that we did everything wrong.
A couple of followup questions:
--Why do you think that the 1160 items come up when you put the site name after site: , but nothing shows up when you search otherwise in google without the site:
--Based on reading the many posts here, how long does it generally take to get trusted
--Big Question---We had several hundred back links at the old site. We are in the process of bring them over to the new site. Is their a possiblity of bring too many of them over too quickly or does the 301 redirect prevent this problem
--Is it OK for a grown man to cry?
--Any thoughts or suggestions that any poster has would be greatly appreciated. I think that this is going to be the most painful financial lesson that I have ever been a part of?
Here might be some tips to help with "trust" rank and speed up indexing. This will also to give you some traffic
1. Put adsense on your site. That way the adsense bot will pull double duty indexing your site.
2. Googlebase/froogle and Google Checkout if your in the US!
1. The 301 redirects take a bit of time before backlink influence gets passed through to the new urls. I'm not sure what that's all about exactly, but I supect it is a kind of trust issue -- Google trying to ensure that this change is not an attempted manipulation. So you are correct to get as many links changed to the new address as you possibly can.
2. Of course, you probably cannot get all your backlinks changed, so here's another issue to watch out for. Backlink influence does not seem to pass through a chain of 301 redirects, but only through a single "hop". So make sure that any other 301s you are doing (such as no-www to with-www) do not create a chain of redirects for a backlink.
This means be attentive to the order in which redirects are applied, and don't redirect from "old-domain-no-www" to "new-domain-no-www" and then to "new-domain-with-www". Instead make sure that "old-domain-no-www" redirects directly to "new-domain-with-www".
3. If you don't already have a Google Webmaster Tools account, do set one up. You don't need to create an xml sitemap, but the feedback that you get about crawling and indexing in those reports can sometimes be useful. It also gives you an authenticated way to communicate with Google, should the need come up.
[edited by: tedster at 1:11 am (utc) on Jan. 2, 2007]
There's one other point I think I should make in this kind of case, especially because I have consulted on fixing several site migrations and redirect issues -- and in almost every case there were problems such as I detail below.
Many times the tech team, web host or registrar tells the site owner that the 301 redirect is in place when it really isn't. There are lots of technical ways to make a request for one url end up showing a different url in the browser. But only one of them is really a 301 permanent redirect -- the server must return a 301 status code to the initial request. I've found that many technical people and server admins are extremely fuzzy in this area, but the search engines most definitely are not.
So accept no substitutes, and always verify redirects yourself. Similar problems come up with 404 response codes. It's only a 404 if the status code returned is a 404 - it doesn't matter what the text displayed inside the brower says.
If you have the luxury of having a url on the new domain that is the exact match for the old url, then make sure the old internal url 301 redirects to the exact internal url on the new domain, and not just to the new domain root.
It doesn't take a rocket surgeon to check this out. One way is just install the liveHTTPheaders extension to Firefox and then start checking. Try all kinds of variations of your urls, old and new aeve completely bogu, and see how the server responds.
And as I said before, accept no substitutes -- or else you may never recover properly from the current bump in the road. Only a 301 is really a 301, and only a 404 is really a 404. If you are not the tech person, then require the tech people to nail it, exactly.
[edited by: tedster at 1:11 am (utc) on Jan. 2, 2007]
I really appreciate your info. THANKS
I would like to cover one question that has kind of been overlooked and is the last one driving me crazy...here it is:
When we search in Google using the following....... site:www.mysite.com, we get 1160 results. But, if you search in google just using mysite, you only get the home page and the other websites where our adword advertisement has shown up.
Why can we not reach our site if site:www.mysite.com reveals 1160 items?
On a general search, the most you will ever see from any domain is two results from any single domain. Putting just your domain name in a regular search bumps into that limit.
One more issue...when we relaunched our site under the new domain, our web designers put meta tegs in that I do not like. The wording is not ideal The new site launched about 4 weeks ago. We lost all of our google indexing and appear to be in the sand box. They have put up 1160 entries, but that is it...we have nearly 50000 products.
So here is my question....is their any danger if I change the wording of the meta tags. Will I lose the 1160 entries? Do you think it will delay getting out of the sand box?
Always test the server responses using an online HTTP header check tool of some sort, or the Live HTTP Headers extension for Mozilla and Firefox. Make sure to test using at least HTTP/1.1 protocol, and be aware that HTTP/1.0 requests often return garbage for features not supported in 1.0 (that's why you test using HTTP/1.1 and need to make 100% sure that the responses are 100% correct).
There are several known issues that catch many people out, and result in the wrong response code being issued. In particular, defining the URL for a 404 ErrorDocument and including the domain name in the URL even if that domain is the same domain as the requesting domain results in a 302 response being issued. The ErrorDocument URL should NEVER include the domain name. Instead it should start with a "/" and give the full folder path to the file. Only then will the correct HTTP status code be issued by the server, and the correct error page be served (a relative URL might also result in the error document itself not being found... so always include the full path starting with a "/").
Changing your meta tags can affect your rankings (positively, or negatively), but it should not affect indexability. The one caveat is to make sure they are all pertinent to the page they are on, not the site as a whole. Otherwise you can incur the wrath of the duplicate content filter.
To confirm that those 1160 pages actually are indexed, search for unique terms contained on the pages within quotes. if the terms occur on many pages, click the link at the bottom of the serps that says "repeat the search with the omitted results included"
moving to a new site and adding 32000 pages was a bad idea.
How can you create 32000 unique pages at once? It's just not possible and not natural and Google knows it.
That is just not true. We added 30,000 plus unique products at one time. If I wanted to, I could have added 80,000. We will have 100,000 products by the end of 2007. My major competitor has 300,000 products on Froogle and over 100,000 in their EBAY store.
The world is a different place today. The least of my problems is adding product.
While I agree with you that Google is penalizing me, you nor Google understand how easy it can be done.
Have a nice weekend
M_Bison is not talking about just "having" that many pages. He/she is talking about adding them at one time. More specifically he/she is talking about adding them at one time to a new site.
It is not natural growth and Google likes natural growth.
Also, Google is not penalizing you. You did that to yourself when you 'started over'. Google does understand exactly how easy it is to create 5 million pages at once, which is why they don't trust your new and improved (and 5X larger) site yet. You have to re-earn their trust, which can easily take years.
[edited by: Powdork at 8:03 am (utc) on Jan. 7, 2007]
A quick question that has me confused.
If I put the following into my home computer, I get 2060 entries in google
If I log on to my work computer from home (or if I am at the office) at the same identical moment with the same identical script, I get 1160 items.
I know that this question is a bit silly, but I am trying to monitor our progress and I cannot figure out what is going on and why they are not the same? I am just hoping that everyone gets the same thing as my home computer!
Your probably seeing different data centers or even different filters based on geo-location. Search rankings are just not "pinned down" in the same way that they used to be.
No, people will not all see what you do on your home computer. In fact there may be many different versions of the results seen around the globe.
|I am trying to monitor our progress and I cannot figure out what is going on |
Yes, it has become quite complex. We haven't even touched on the personalized results that people can get when they are logged in to Google.
That makes sense...thought I had lost a brain cell....Thanks