301 only pass 100% PR when it’s close match

Selected answers from the Dumb SEO Questions Facebook & G+ community.

• 
  Michael Martinez: First of all, don`t ever do a redirect "for SEO". Put the user experience first. You don`t know if the redirect will ever pass value. Googlers have said in the past that they can deprive a site of its ability to pass or receive PageRank for bad behavior. On the basis of that, it`s reasonable to assume they have very granular control over what is allowed to confer or receive PageRank.
  
  Secondly, if you`re feeling that deep a need for PageRank, it would be better to look at why you`re not earning the links that should provide it. Internal linking is important. And there are good reasons to remove or consolidate content - even if it`s "for SEO". But at the end of the day you can only give youself so much of a boost.
  
  PageRank represents a mapping of value across Google`s index. They want it to flow freely from document to document and they recommend using good navigational structures that make sense to people. They want to follow those navigational structures, not compensate for a lack of them.
• 
  Shao Chieh Lo: Thank you Michael Martinez, very insightful
• 
  Roger Montti: I`m not sure if it`s an all or nothing deal with 301s. I suspect it is.
  I think I wrote that article and if I did then I quoted everything he said about it.
  
  I don`t leave things out except maybe when he says, "I think" or repeats words like "so, so, well, well." That`s why you`ll see three dots "..." that`s me cutting out the pause phrases.
• 
  Shao Chieh Lo: Roger Montti thank you so much for explaining!
  
  I agree, I think what John means it`s an all or nothing thing.
  
  However, it`s also possible that it`s not all or nothing but John can`t addmit it since it will open up the door for internal link manipulation I guess~
• 
  Roger Montti: Shao Chieh Lo
  All links pass a depreciated amount of PageRank. It`s not only depreciated by the amount of links on the page, but also by a small amount that represents a decay in the links power as a citation.
  
  301 redirects USED to pass a depreciated amount of PageRank in order to carry over that decay. The fear at Google was that people would 301 redirect their navigational links in order to pass 100% of PageRank.
  
  So in terms of SEO, everyone used to avoid doing 301 redirects for fear of losing some PageRank in the process.
  
  However, there was a change at some point, announced during the transition from http to https that said there would be a one to one passing of PageRank, with no diminution of PageRank. That way, when you switched to HTTPS you wouldn`t unnecessarily lose part of your PageRank. It would be a one to one match.
  
  As part of this change, the terms of that PR passing is similar to what goes on with Canonicals. That the two pages must substantially be equal in order to pass the full amount of the PageRank.
  
  If they were not equal then the page being redirected would be seen as a soft 404. I am pretty sure that`s how it works.
• 
  Shao Chieh Lo: I agree, I think I will not redirect the informational page to product page then, it can be too risky
• 
  Roger Montti: Shao Chieh Lo
  A typical strategy is to create an informational page and embed a link to the product page from within the content. That way all inbound links to that info page will be just one click away from the product page.
  
  I believe it`s most effective to add that link at the time of publication. However info pages are frequently updated so probably no harm in updating that info page with a link (as long as there is context).
  
  It would be great to attract links straight to a product page but that doesn`t always happen.
  
  The strategy that some still cling to, of getting links straight to the home page is seriously outdated, IMO.
• 
  Richard Hearne: Roger Montti "That way, when you switched to HTTPS you wouldn`t unnecessarily lose part of your PageRank. It would be a one to one match."
  
  The question begging in my mind is whether this change was applied to a strict set of circumstances, such as HTTP > HTTPS migration, and not more broadly to everything? Some myths have a habit of starting with an announced change that somehow becomes applied to a broader set of cases than is actually announced. Any thoughts?
• 
  Roger Montti: I don`t know if it was done specifically for HTTPS migration or if it had already been that way and they hadn`t told us until that event. I suspect it had already been that way and they hadn`t told us.
  
  It sounds similar to how canonicals work in that it has to be close to a one to one match for the canonical to work.
• 
  Richard Hearne: Roger Montti Thanks. I`m much more curious about decay factor. If there is none than the same amount of PR would pass across 3 redirects as 1. It`s something I`ve long wondered about, but nigh on impossible to know.
• 
  Roger Montti: Richard Hearne That`s a chained redirect, something that`s always been recommended against.
  
  In the distant past I remember it caused a site to not show any PageRank in the G Toolbar. A site had migrated from several different technologies like CFM etc and had chained redirects.
  
  So I emailed someone at Google who recommended unchaining those but he also fixed it so that the PR showed appropriately. Maybe just a bug in the toolbar? He never said.
  
  I was invited to Yahoo HQ and during the visit one thing the person there said to me was keeping things simple was always the best approach. It will minimize the possibility of a crawler/indexing error. I think that remains true.