Forum Moderators: open
www.mysite.com?affiliate=Bob
www.mysite.com?affiliate=John
www.mysite.com?affiliate=Dave
Will such links contribute to PageRank of www.mysite.com?
I have a site in which a certain well-linked page (banned by robots.txt) has a PR of 3, but every link to it has a different variable in the querystring. If you visit the page with any of the major variables in the querystring, you get a pr of 0. I also have lots of dynamic pages (again differentiated by the querystring) indexed.
So yes, these links will contribute to PR (in my opinion anyway). However they can cause problems. In particular, occasionally affiliate links appear in the serps before your own. So if you possibly can, its best to run these kind of things with server side referrer- detect coding instead. Admittedly that sort of coding is not infallible, but it works most of the time.
Not such a bad deal!
Websites that run affiliate programs need to look at their affiliates as allies -- their own sales force -- not as competitors. Some of the biggest affiliate networks have this backwards.
If you use a home-grown affiliate program, the affiliates link directly to your domain, and Google might be able to track these links.
I do feel strongly about affiliates though. Your link will usually appears on higher traffic pages and in more relevant locations than the never visited "links" pages.
Now coming to "Do affiliate links contribute to Pr".I was wondering the same.We as a company have 1500 affiliates, but most these point to index.htm page where as the page that is spidered by Google is index.html.
For me it's a clear cut no, they don't contribute to a large extent but sometimes if an affiliate completely puts a page on his site then perhaps it might be indexed.
Personally I feel affiliate url is good way to measure a site.The more number of varied Affiliates and high pagerank affiliates the orginial site should have a good pagerank, since the concept is why would lot of people will be affiliates for a site, since it pays and is generating enough revenue.
I hope Google takes this far more seriously as way of link analysis.
Aravind
For example, if a site links to www.yoursite.com/affiliate.jsp
read the referer domain by doing a request.getHeader("Referer") Use this domain name to identify the referrer instead of using the traditional? parameters. That way google will not be treating these links as referral links.
On my sites I put in a NOINDEX,FOLLOW robots tag when a SE spider fetches an URL with affiliate ID, and I provide clean links instead of URLs with tracking codes. I hope this kind of cloaking won't get me in trouble with Google, because it avoids useless indexing of duplicate content. GoogleGuy has stated that cloaking is allowed to truncate session IDs in URLs for Ms. Googlebot. I think changing meta tags and cutting affiliate IDs from internal links in this case is at least pretty similar.
Yes, if implemented correctly and policed.
I'm not big on affiliate marketing, it pays for a few dedicated servers, but I never see it going much further.
There is a fundamental flaw with affiliate marketing on Google. If the affiliates are "good boys" and link as requested they end up killing themselves eventually due to the Google Algo. So a smart affiliate will deliberately buck the trend and break the "rules". While you get away with this there is hope.....but long term I see problems on the horizon.
How about the affiliate that implements the link in JavaScript, PHP, PERL or ASP or another mechansm so that the link can't be followed?
All of the above can be used to prevent the supplier from gaining unduly from affiliates....how many people apply this twist today?...not many. Which is why I said that long term most affiliates are on a road to self destruction!
Am I correct?
Will such links contribute to PageRank of www.mysite.com?
It would depend on what happens at www.mysite.com?affiliate=Dave. You could use ASP, JSP, Perl, PHP etc. to set a cookie or a session variable, and then 301 or 302 redirect to the actual page. This way google would follow the redirect and apply the link www.mysite.com?affiliate=Dave to wherever the link was.
I would avoid however getting www.mysite.com?affiliate=Dave to actually result in a page as then every Dave, Bob and Henry will count as a separate page.
IMHO.
So, if I have three affiliates, and their links (each with the corresponding parameter "affiliate") were followed by the Googlebot, Google will see three separate pages that have identical content. Right?
But SebastianX has experienced the opposite -- see his post #12 in this thread, on the page #1.
Giving Googlebot a permanent redirect (301) should be the best solution to gain incoming PR, but you can't track no-cookie traffic, except when you redirect SE spiders only.
Could you not set a session variable, thus:
(in ASP/Visual Basic, as that is the only language I am conversant with)
<%
Session("Affiliate") = request.querystring("affiliate")
Response.Status = "301 Moved Permanently"
Response.addheader "Location", "http://www.mydomain.com"
Response.End
%>
Wouldn't this add a session ID to the URL when the user has cookies disabled, and robots?
I'd check the UA for SE spiders first, ignoring the affiliate ID in this case. And I'd ensure spiders get the same content as surfers. I'm not sure whether Google tolerates this method.
Absolutely, but you have to keep the tags simple. index.html?1234 seems to work consistently while index.html?shyfdtgsrefvvdongyshr__yhstevbfllsgskflhenm doesn't.
if you are the person getting the links, you want to keep them as simple as possible to benefit you. If you are the one giving the links you want them to be as simple as possible because you are passing along more value to the parent site than merely the traffic. Pagerank may mean little these days, but it means more than zero.
www.mysite.com?aff_Bob
www.mysite.com?aff_Dave
www.mysite.com?aff_Henry
And don't use variable=value pairs like this:
www.mysite.com?affiliate=Bob
www.mysite.com?affiliate=Dave
www.mysite.com?affiliate=Henry
Which one is better? Or Google simply prefers shorter URLs regardless of the contents or structure of the parameter string?
I have observed that my competitor gains a lot by this.
Since the spider is directly conneted to www.widget.com/index.html , it will definetly take this url into account.
the moment it detects "?" it will ignore what comes after "?" am I right!
or is it like the moment it detects "?" it will completely ignore the URL.
Could you clarify on this?
Aravind
www.mysite.com/news.asp?article=may10
www.mysite.com/news.asp?article=may11
www.mysite.com/news.asp?article=may12
So why wouldn't it count the following as seperate pages?
www.mysite.com?affiliate=Bob
www.mysite.com?affiliate=Dave
www.mysite.com?affiliate=Henry