 |
||
 |
|
|
Internet Marketing Coach
Internal Linking and PageRank Leakage
September 11th, 2007 | 13 commentsIf you’ve been online for more than a day, you know what Google PageRank (PR) is. You may also know that PR is calculated on a page by page basis rather than on a site by site basis as well. Many people are confused about pagerank, thinking it’s worthless. The PR that so many people say is worthless is the toolbar PR, but that’s a whole other story that I explained in a post titled Google PageRank and Alexa Ranking Explained.
What I want to talk about right now though is your internal linking site structure and pagerank leakage. What does your site structure have to do with your pagerank? It actually has a lot to do with it.
First you need to understand what “pagerank leakage” is and what it isn’t. Let’s say you have a page that has an actual PR of 6. That page has 30 outgoing links, both internal links (other links from your site) and external links (links going to other sites on the web). That PR6 page now distributes all of that PR to each of those links. This does NOT mean that the PR6 site is losing pagerank. It’s pagerank will remain. So, linking out to internal and external pages is NOT what PR leakage is!
Hopefully that PR6 page isn’t the only page on the site so now you need to think about the page as a small part of a network. That page, and every one like it, has PR to distribute. If all of these pages are distributing boat loads of PR to external pages (not your own pages) then you are “leaking pagerank” because that “link juice” could have been recycled back into your site making it stronger as a whole.
I’m not saying you should never link to other sites on the web, actually you should be. That’s what the world wide web is all about, and if you’re not linking out you’re not participating in the WWW. What I am saying though is to limit your external links. If you find a good blog post that is related to your post/page then by all means link to it.
Here is where so many people go wrong (including myself until Andy showed me the light). Every single link on your pages receive a percentage of the PR assigned to that page. That means all of the following are taking PR away from your site!!
– Social Bookmarking links (add to google, digg this, etc.)
– Contact pages
– Affiliate Links
– Search Results
– Print Pages
– Blog Rolls (Have you seen the size of some people’s BlogRolls? MANNNN that’s alot of sitewide PR leakage!)
How to prevent PageRank Leakage
Very simple, rel=”nofollow”
Add the rel=”nofollow” attribute to all of your worthless links that shouldn’t be taking from the PR that could be recycling back into your site.
For those who don’t know how, here’s an example:
Normal link: <a href=”http://ez-onlinemoney.com/”>Make Money Online </a>
Normal link with rel=”nofollow” attribute:
<a rel=”nofollow” href=”http://ez-onlinemoney.com/sitemap.html”>Sitemap</a>
Not only will this preserve PR, it will improve the spidering of your site by giving the spiders less to do, pointing them to the money pages.
Now get to work fixing your 5,000,000 sites and when your done come back to leave me a comment 
Popularity: 11% [?]
Related Posts
- Internal Linking - META nofollow, rel nofollow, robots.txt Confusion
- I’ve Changed The DoFollow Comment Policy
- Free Blog Review #3 - ThomasSinfield dot com
- 6 Underutilized Ways to Increase SE Rankings
- Google’s up to it Again - More Toolbar PR Reductions
- 11 Blogs that Accept Guest Bloggers
- What’s That Link Really Worth?
- Duplicate Content - It May Not be so Bad
- 5 Ways to Benefit from High Google PageRank
- Does Your Site and/or Blog Need a Sitemap?
Twitter
Linkedin
11th September, 2007 at 9:00 am
For ease of use when talking about HTML there is a short hand whereby you type the “and” character followed by lt or gt and the semi colon. The letters stand for less-than and greater-than which is what we nick for HTML pointy brackets. I have no idea if your comment system will escape my escapes (it should) so you may see two pointies or the raw code… <a href=”stuff”>link text</a> - although it looks a mess when you type it the screen should show the characters you want but not treat it as real HTML (even though if you copy and past it the result can be used. Tricky stuff, no.
11th September, 2007 at 9:07 am
OK so you allow HTML, cool. Unless I have cocked up this comment should show what I typed to show you what I would type if I was going to type a hyper link… <href=”stuff”>link text</a> If not you might want to delete my comment.
On the subject of page rank I have my doubts as to how much use nofollow is. Some pages should be followed but not indexed. For example with a site map (something that is good for internal PR) the robots=”noindex,follow” meta tag might be a better option. Searchers do not find the map but the engines do and they follow it out again.
I would also use nofollow on social bookmarking links but definitly do follow for comments. The reason for my rarely using nofollow is that once other bloggers see that I am keeping the link juice to myself they will wonder if I am worth links with juice and suddenly I have no links in and no PR to save.
11th September, 2007 at 9:58 pm
Hey Matt, thanks for stopping by. I just bookmarked your blog yesterday. I’ll have to look in to your html suggestions a little more, new to me. Of course I’m not html expert by any means.
As far as the nofollows. You’re absolultely right in the fact that you don’t want to keep all the PR to yourself. I use Lucia’s LInky Love plugin for comments, rewarding regular contributers with dofollow and nofollowing one-time commentors. I never nofollow legit links in my blog posts. If I find a good blog post it absolutely gets a PR passed!
My point above is really to be sure and nofollow unneeded links like social bookmarking sites etc.
Sitemaps I believe should be left alone with no nofollow attribute at all.
27th November, 2007 at 2:42 pm
[...] annoyed when their posts and/or pages drop in PR accordingly. Oh well, at least you won't be leaking any PR from those [...]
18th March, 2008 at 3:49 am
Thanks Josh - New to the fine art of keeping the juice. What does a rel=”me” do to page rank. For example if I’m linking to my social bookmarking sites, but I can’t put a no follow because I already have a “me” stamped in there… am I leaking?
Thanks for the input
18th March, 2008 at 4:07 am
Do the BlogCatalog widgets (or any widget for that matter) that links back to an external soak up pagerank juice? All those little lines of scripted code that I insert - do we have some leakage going on?
As well, since there’s no visible links to modify, is there a way to prevent juice leaking from javascript?
Thanks. Cheers.
18th March, 2008 at 3:27 pm
@ Web - the rel=”" tag is called an XFN tag. I don’t know a whole lot about it, but it’s a way of telling the search engines the relationship between the site owner and the site that is being linked to. It is not a nofollow, so if nofollow isn’t inserted, juice still flows. You can read more about it here: http://www.gmpg.org/xfn/
The googlebot does not read Java, so there is no leakage going on with java widgets.
18th March, 2008 at 4:44 pm
Thanks for the response, Josh. Checked out your lead on more info on XFN. Appreciate it. I’ll probably remove the rel=”me” tags and replace them with nofollows. I’ve signed up for your rss feed to keep in the loop. Thanks.
1st April, 2008 at 9:51 pm
Another great article Josh. Have been letting a lot of juice flow from my sites. But not anymore. Going back to fix them now, so thanks for the tip.
Steve
26th May, 2008 at 4:19 pm
[...] of you who have been following this blog for a while know that I’m big on developing good internal linking structures. A good internal linking structure can not only increase the number of pages your site/blog has [...]
26th May, 2008 at 4:28 pm
[...] useless pages will not. To ensure the right pages are indexed be sure to use robots.txt and a good internal linking structure. If you’re running a WordPress blog there are many ways to prevent duplicate pages including [...]
26th May, 2008 at 6:14 pm
[...] Examine your internal linking structure. Nofollow all links that don’t deserve authority i.e. contact, email, affiliate links, social [...]
4th June, 2008 at 9:55 pm
Josh,
I have just found this post.
I tried to add the no-follow attribute to my links, but I have got a web 2.0 joomla site which does not allow you to do this in the main body, only in the modules, which is very annoying.
Barb.