| 5:25 pm on Jun 24, 2010 (gmt 0)|
It was confirmed recently that 301 redirects bleed some PR. I believe also that Google can get lost following too many redirects, but I don't know if there's a hard and fast number. In general I believe that current thinking from many is that redirects should be used when all else fails.
| 8:12 pm on Jun 24, 2010 (gmt 0)|
thanks for your reply. Unfortunately it doesn't answer the key question., which is, will my site or the redirected site benefit from the links?
| 8:19 pm on Jun 24, 2010 (gmt 0)|
FranticFish did answer your question. ;)
|It was confirmed recently that 301 redirects bleed some PR. |
What the above means is that redirected links to your site are deprecated, they lose part of their value. What FF posted is your answer. Redirecting links causes their link value to drop.
Then FF elaborated further, beyond your initial question (a bonus bit of knowledge), that multiple 301's can cause issues, that simpler is better. Thanks FranticFish for going the extra mile on that answer to make it more complete. ;)
| 9:04 pm on Jun 24, 2010 (gmt 0)|
MB: thanks; I probably could have been a bit clearer. I know what I mean, even if that's not always what I say :)
WC: Curious as to why you don't want to get the links direct. Is that not an option because the content is different?
I've not done loads of 301s, but the last one I did (two years ago) was with identical content from one domain to another (rebrand). The new site ranked where the old one did after about 6 weeks. I'm aware that it can go smoothly or very bumpily even when the content doesn't change and that some sites take a hit for months.
I can't imagine why you'd want to redirect unless the content is different. You might find that Google has ways of discounting the IBLS to Page A altogether if it redirects to Page B which is different, because if this is a trick you're planning, it's one that has been used extensively.
If this is what you're planning, you might find it a waste of time.
| 2:12 am on Jun 25, 2010 (gmt 0)|
Thank you both for your replies. FF, in answer to your question, my interest lies in control of the links. There may be a situation where I would want to remove the links and the only way I could think to do this was through a redirect.
If you know of any other way to control links, I'd like to hear from you. Thanks again.
| 5:03 am on Jun 25, 2010 (gmt 0)|
If you've read the white paper from Sergey Brin and Larry Page (Google founders) while at Stanford called The Anatomy of a Large-Scale Hypertextual Web Search Engine which became the blueprints for Google and the first place where the idea of (and original calculations for) PageRank was discussed... and if you read the Google patents related to PageRank... then you'd know that there is something called a damping factor in the PageRank formula which causes a natural decay or leakage of PR with each hop or link.
Though the formula has obviously changed over the years, this idea of a damping factor likely still remains. Originally, it was set to about 15%. A VERY simplistic view of PageRank is described in the white page mentioned above.
Basically, if a page has X amount of PageRank and Y outbound links then each outbound link is passed approximately X/Y PageRank. I say approximately because it's actually ((1-D)*X)/Y where D is the damping factor (for example D=0.15 if the damping factor is 15%). So the total amount of PR the page has is first reduced by some amount based on the damping factor and then divided by the number of outbound links on the page to detemine how much PR is passed out.
If you think of a 301 redirect as basically the redirected URL having a single outbound link pointing to the target URL of the redirect then it would stand to reason that it is this damping factor that is the cause of the decay or leakage in PR.
For example, if the page being redirected previously had X PageRank, Y=1 (because there is only one outbound link of a redirected page - the target of the redirect) and the damping factor D is, say, 15% then only ((1-D)*X)/1 PR is passed to the target of the redirect... or 85% of the PR of the page being redirected is passed to the target of the redirect since the formula above reduces to ((1-0.15)*X)/1 and ultimiately to 0.85*X.
My experience has been that Google is actually VERY good at following multiple or stacked redirects... even stacked redirects that are 3, 4, 5 levels deep. However other search engines might not be as good.
Stacked redirects should ALWAYS be avoided if possible because of this damping or decaying of PR with each hop as well as the fact that all search engines might not be as good as Google at following them.
Instead of a set of stacked redirects like:
You should use individual redirects all pointing to the final destination:
to maximize the amount of PR passed to URLD.
As FF was saying, it's inefficient in most cases to build links to URLA only to 301 redirect them to URLB. You're better off simply building subsequent links directly to URLB after URLA has been redirected to URLB.
The only time I can think where it would be advantageous to build links to URLA after URLA has been redirected to URLB is if you own URLA but you do not own URLB... and you think that at some point in the future you might want to stop redirecting those links from URLA to URLB and redirect them to some other URL... possibly one that you own.
| 5:35 am on Jun 25, 2010 (gmt 0)|
Thanks very much for this extremely informative post. This is exactly the information I was seeking and the situation you describe in your last paragraph is almost 100% correct.