Forum Moderators: Robert Charlton & goodroi
Here is my experience with a recent 301 redirect:
--Popular travel niche website approx 4 years old.
--Actively covering all topics related to my niche but we also sell entire vacations… think of this site as being a vortal covering everything and anything dealing with this niche including up-to-date news, weather, unique articles, forums, interactive tools for planning a vacation and a bunch more all of which are free.
--Very little link trading with the bulk of links coming in naturally
--Very little outbound linking
--Clean HTML (for the most part)
--Some JavaScript but nothing black hat or meant for SEO
--Listed in DMOZ, Yahoo Directory, Zeal and Google Directory
--Was a PR 4 with about 50 inbound links
--Index count was 6,080
--Was in the top 10 results pretty solid even through Bourbon and other various updates
The 301 bomb (website suicide), applied a domain wide 301 redirect via IIS to a domain that is 18 months old. E.G. olddomain.com/widgets --> newdomain.com/widgets
I have seen some people post “why would you do this?” … this isn’t a valid question in my opinion because there are lots of very good reasons to do so.
--301 was put in place roughly 80 days ago
--After approx 5 days the site was nowhere to be found in the SERP’s
--Sent a request to help@ and was told the site was not banned or penalized
--Started the long waiting process
--Quasay non existent update Gilligan started
--Old domain was stripped of PR across all DC’s
--New domain still has no PR on any DC’s
--BL’s update to 138 on most DC’s
--Google Directory updated showing the new domain as a PR 6 and at the top of my niche
--site:oldsite.com would reveal the new domain
--index count is fluxing between 10,300 and 10,900
--PR begins to return to the old domain!
--alas, no where in the SERP’s even after going 50 pages deep.
Sounds like classic sandbox in my opinion but I think a better name would be “Gilligan’s Island” because most of us in 301 club feel stranded on a deserted island with no hope of rescue but occasionally there is a glimmer of hope on the horizon.
I also want to point out that until you have been through an experience like this it’s not helping anyone to call people in this situation whiners, or something inflammatory because we are simply trying to figure out how to make a some what smooth transition and to avoid the sandbox.
Well, if you are still reading you are probably in this position now but if you are thinking about doing a 301 redirect, do so understanding that you will loss rank for at least several weeks.
Here some alternatives that have been discussed
1)Meta refresh to new domain – bad, could get a dupe content filter
2)JavaScript redirect – bad, looks too black hat or spammy
3)302 redirect – is not permanent and is also very spammy looking
4)404 all old pages – don’t know how this would work
5)Build a new site which simply wasn’t an option for me because I have a lot of unique content that would take weeks to regenerate without having any duplication
Another way to look at this was put best by jd01
It appears...
New Domain with 301 from old site = New Site
New Domain with no redirect from old site = New Site
New Domain with meta refresh from old site = New Site
New Domain and old domain with old content = New Site & Dup Content
IOW New Domain = New Site
Don't change if you don't have to - the, for lack of a better term, sandbox is in play.
Justin
Being that GoogleGuy is the closest thing we have to a direct contact (for most of us anyway) I would greatly appreciate his feedback.
What canonical page fix?
302 hijack maybe fixed (well less obvious to see) but canonical page problems are rampant.
[google.com...]
As this was the official advice I did a 301 at the start of August. After that it was the familiar story of being dropped after 5 days.
I'd love to hear from GoogleGuy about this problem. It appears to be affecting a lot of legitimate websites. Modemmike's idea of a paid review service may be a way that Google could deal with the problem. Maybe if the mods put this thread on the Webmasterworld home page it might get the attention from the people at Google that it deserves.
BTW, if Google does read these forums then they surely saw this thread but I suspect there is too much room for manipulation via 301's so nothing will ever be done. My gripe is why they don't say so in the guidelines... you know, a simple, "changing a website domain name will be treated as if the website is entirely new", (only more wordy).
One way of knowing when Google is unable or unwilling to fix the problem is when the problem is not being fixed.
My follow up comment is already there too :)
Will my rank for those pages eventually return to the former ranking on the original domain, or does the 301 permanently connect the old URI to the new URI?
I initially denied the sandbox theory as being a factor here but that was because I was in the denial stages of my grief... I now just want to know what causes sites to be released from the litter box... is it after an update, or during a mass exodus or what?
GoogleGuy, can you offer any insight into this topic?
We have had to move servers many times in the past few years and each time had to redirect with the server number in the URL. Our actual domain name has never changed.
To make things even worse, we are required by the server admin to use 302 redirects (301s are not allowed) from the old servers to the new ones.
Here is a short history:
1st URL: [site.com...]
2nd URL: [www-123.site.com...]
3rd URL: [www-456.site.com...]
We are now listed in google for [site.com,...] [site.com,...] [www-123.site.com,...] [www-456.site.com,...] etc.
This was not a big problem until Bourbon. Now all our results in Google are a mess and our traffic from Google has taken a nose-dive. We are at a loss as to what to do.
And that is with a 302 redirect. Now I am not even sure if a 301 would help us.
And from what I am reading here -- we can't even ask Google to remove any of the old URLs with server numbers because it is all the same domain. And we don't want to remove the entire site.
Sorry to go on and on in my first post. I am a good whitehat SEOer and I have been doing this for many years. This is my first seemingly unfixable problem in Google.
The company I work for owns the servers. So we can't move to other servers -- because it is our company Web site. We are currently fighting to get these 301s in place of the 302s. I am just hoping that if we do succeed with the 301 battle, that we will see the results -- that is, our Google results will be fixed to the correct URLs and our clickthrus from Google will increase (or return to close to where they were before Bourbon).
We once had 301s, and then corporate policy changed and we were only allowed 302s.
Talk about inane bureaucracy...
You just said "We used to be able to define our redirects (according to HTTP Protocol 302 is undefined -- 307 is temporary and 301 is permanent), but now we can't, because our company made a decision."
Glad I can refuse to work for anyone that stupid...
Good luck.
Justin
The truth is, most people in a position to make a decision don't know or care about the technical details. I'm guessing some slacker in your tech department couldn't be bothered to work out the details on an issue that could have a signifcant imact on your business.
Recently remembered about one of the domain names that I registered last October. 1 Page Website(was and is now.) Has, one large graphic and about 150 words of text and a keyword reach link to it self on the bottom of the page (A domain name with spaces between words). Page got pr2 after Bourbon Update and Ranked at 46 for the phrase. 3 weeks ago put a clean 301 from none-www to www. and put a date stamp on the bottom of the page.
Here is the thing.
Type domainname.tld or www.domainname.tld in search box. Result is the same as it should be(cached version returns 2 days ago cached copy).
Type the keyword phrase site (tanked 203 points to 249) - cached version is from before I did a redirect (has no date stamp on it). "Site:" command has the old version as well.
I don’t believe that putting a date stamp on the page could do such damage...
It's been about 6 months now sinse the websites all tanked.
It had to be the 301 redirect because all of the websites were different content and built differently but they were also similar in their html. They all had H1 tags for the header etc...
But the 301 was the only thing I did to all of the sites at the same time. Then a month later they were all gone. Completely gone.
Your search - site:www.mydomain.com - did not match any documents.
That's what I get.
The down side is that Google is now re-showing (as of a few weeks ago) loads of www pages, and all of them with a cache from a year ago, instead of hiding the www pages away as asked (by the 301 redirect).
I think the problems arise when time has passed after the launch of a site and then a 301 is put in place. After a site is launched, Google likley has pages of both www and non-www (because of different ways links point to a site regardless of how you link internally within your site). If Google has both www and non-www pages at some point in time, and a 301 is put in place, it seems like Google just can resolve that properly.
Just a theory not based on much beyond speculation. Any thoughts, any evidence to support or disprove this?