Calendar
Network
Feeds
- 43
- Sphinn It!
Went Hot: July 18, 2007 - 4:41 pm
Posted By: vanessafox 717 days ago
Topic Type: News Story (Jump to http://www.stonetemple.com)
Category: Google SEO
#9 mentions that at SMX, I was asked if Google penalizes sites that make heavy use of nofollow on internal links throughout the site. Eric is right in summing up my response, but I would add that while Google doesn't care how many links are nofollowed, the reason I suggested using robots.txt or a meta robots tag to block the targets of those links was that nofollow isn't a great way to tell search engines not to index a page. You might nofollow the link to the page, but another site is likely to link to it and all of a sudden, like magic -- it's in the index.
Another tip I would add that he doesn't have listed is that you want the canonical version of the URL to have few parameters if possible. URLs with lots of things that look like session IDs and things may be less likely to be algorithmically seen by search engines as the version to index.
7 Comments
Comments
Vanessa,
"URLs with lots of things that look like session IDs and things may be less likely to be algorithmically seen by search engines as the version to index."
That sounds very interesting indeed! Would you be kind to elaborate more on that, when you have time ;-)
Sure, it's just that one situation I see duplicate content come up a lot is with dynamically generated URLs. And Google's guidelines have always suggested using as few parameters as possible and not using parameters that look like things with session IDs. If two URLs have exactly the same content and the URLs look like this:
mydomain.com/products/items?id=123
mydomain.com/products/items?id=123&session=456&browser=ff&returnvisit=0
(or whatever)
The first URL would tend to logically look like the canonical one. So I'd just look at the version of the URL that you want to be indexed and make sure that your content management system isn't adding a bunch of stuff to it that it doesn't need.
Vanessa,
Thanks for feedback.
It seems you have just changed your avatar while I'm writing my comments.
http://sphinn.com/avatars/user_uploaded/vanessafox_50.jpg
Looks great :-)
I consider the nofollow and robots.txt as being used complimentary to each other within a site. I think of it as a courtesy for the bot. Let's say I don't want my feeds indexed, because I hate clicking on a search result and getting sent to feed, so I block them from being indexed with robots.txt. I'll also nofollow links to them, because it seems kind of dumb to tell Google not to visit a page, then show them a link to that page. To me its the same logic as not having links on your site to a page that redirects to another page. The redirect may be there for visitors or bots, but why tell spiders to crawl the link when you know its going to be redirected? How many sites have you seen the have the "home" links all go to "/index.php" which is then redirected to "/"?
Is there a benefit? I'm not sure, other than I am using less resources of the crawlers, leaving less to chance by an algorithm making the right choice, and I'm more in charge of what does/doesn't get indexed.
Then again I think different about nofollow than most people. I'll actually give people who comment on my blog link credit when they write something that is published on my site.
"Speaking URL's" are better from a user experience perspective too. URL strings are an overlap of usability and SEO and as a general rule of thumb, by improving the usability of your site you're also making an improvement to the site's SEO or findability. Speaking URL's just make sense, and that's why they're a good habit to get into...
I would never use "nofollow" on internal links within a site. That sends a very confusing message to bots.
You have the choice of a meta robots noindex tag on the target page itself to keep it from appearing in the index at all, or the use of the robots.txt file to stop it being spidered (but that doesn't prevent it appearing as a URL-only entry in the SERPs).
See also: http://www.stonetemple.com/blog/?p=169