Over the last eight months or so of offering SEO consulting services to clients via Outspoken Media, I’ve seen a lot of interesting things. One of the more common (and surprising) things we see are URL structures in “current” sites that are completely “root based”. What does this mean?
Back around 2003 (hey I’m old, I don’t remember the exact date), toolbar pagerank was all the rage and most optimizers were all about creating those green pixels and maintaining it through any means necessary.
One of the things optimizers noticed early on was that Pagerank seemed to be site based and distributed from “the top down”.
What this means is that if your homepage was a low to mid PR5, any pages directly off the root of the homepage would be a PR4. Subfolders directly off the homepage would also be a PR4. But pages within the subfolders would be a PR3.
yourdomain.com = PR5
yourdomain.com/your-page.html = PR4
yourdomain.com/your-folder/ = PR4
yourdomain.com/your-folder/this-page.html = PR3
yourdomain.com/your-folder/another-folder/ = PR3
yourdomain.com/your-folder/another-folder/another-page.html = PR2
Back then, the higher your toolbar PR was, the higher the likelihood that your pages would rank (that wasn’t all there was to it – lower PR could beat higher PR, but toolbar PR was a big portion of things in those days). So a lot of optimizers took to creating “root based” sites. Which essentially means that every page of the site was built off of the root.
yourdomain.com = PR5
yourdomain.com/your-page.html = PR4
yourdomain.com/your-folder-now-a-page.html = PR4
yourdomain.com/your-folder-now-this-page.html = PR4
yourdomain.com/your-folder-another-folder.html = PR4
yourdomain.com/your-folder-another-folder-now-another-page.html = PR4
Back in the day it made sense. Nowadays, it’s pointless and messy.
First things first, Toolbar PR is for entertainment. Most optimizers have known that for a long time. Secondly, PR either didn’t remain or never was “site based” (you can decide that one) and is instead based on the individual page factors (like inbound links). Thirdly, the Google algorithm is no longer a one ingredient wonder. Nowadays, we have trust, age, authority and varying other factors contributing to how a site ranks in addition to Pagerank.
In addition to the strategy now being pointless, it’s also messy.
Not only do you risk getting crawlers confused with a lack of a logical site structure, but it is also beyond annoying to someone who has to work with a site to have to trudge through a root based site with several hundred or thousand pages.
It’s like opening a huge walk in closet to find a tie only to not be able to even walk into it because everything is simply thrown on the floor with no rhyme, order or reason.
But it seems a lot of “SEO firms” didn’t get the memo and we are constantly seeing sites with this “SEO method” employed.
Cleaning things up
So now that you know this “method” of SEO is archaic, ineffective and sloppy, how do you go about fixing your site?
Whether or not you should “fix something that isn’t broken” is something only you (if you have the knowledge), your in-house SEO or outside SEO firm can really answer as it really does need to be looked at on a case by case basis. There IS POTENTIAL RISK INVOLVED with changing URL structure that should be assessed. That said, I’ve had a lot of success with site structure migrations and most times, will choose to slowly migrate the site to a new, sensible URL structure.
Some tips for site structure migration
If you do choose to change your URL structure, you’ll find some tips based on my experience with my previous migrations below:
- Prepare for the fact that it takes Google a bit to “figure things out”. The more often you get crawled, the less time it will likely take for Google to get with the program. I’ve seen URL migrations take anywhere from a few days to a few weeks to get sorted out.
- Always choose a small, “mid-level” traffic section to start with. This way, you can see the results before enacting things on a larger scale and/or with your most important keywords.
- Make sure you have a good 301 plan in place for pointing the old URLs at the new ones. Without it, your new URLs will not “take the place” and the authority, link popularity and the rankings of your old URLs.
- Change site navigation, but use your internal sitemap as a “reminder”. What I mean by this is that I usually change the links to the new links throughout the site naviagtion and within the site content, but I leave a link pointing to the old URL on the internal sitemap (not the one you feed to Google via WMC) until I’m sure Google has seen the redirect and removed the old link from the index.
- Wait a few days, watch for the new URLs to be indexed, the old URLs to be removed and wait to ensure the new URLs take over the rankings formerly held by the old URLs.
- If everything goes smoothly then I wash, rinse and repeat with other sections. Take things slow. It takes a while, but if anything goes wrong, you want it to go wrong with one small piece and not your entire site.
In addition to the above tips, it never hurts to go out and get some quality links to the new URLs from tightly themed sites to let the engines know that your new URLs are just as relevant as your old ones and that their still important to crawl regularly (and hopefully speed up the migration process).
About the Author
Rae Hoffman is the CEO and Co-Founder of Outspoken Media as well as the author of the often controversial Sugarrae Blog.
Get social with Rae at Sphinn | Twitter | WebmasterWorld
Share this Post
Stay Connected
Subscribe to the Outspoken Media feed via RSS or email and follow Outspoken Media on Twitter!
{ 11 comments… read them below or add one }
What do you think of the idea that Google allegedly gives preference to words that appear earlier in the URL structure as opposed to later?
Do you think adding a structure – like on Outspoken with domain/category/post-title – will hurt your chances for ranking for post specific keywords, but help you rank for category specific keywords?
Cool… the related word search can be a big help in determining logical paths i.e folder/page names.
I envy you for having clients who have made it to 2003 ;)
I’ve been doing a lot more small-business consulting in the past couple of years, and I’m amazed how much overoptimizing I see. People hear one SEO tip (usually outdated and with no context) and blow it completely out of proportion, basing their entire site around it. Subtlety is completely lost on them.
Hi Rae,
Looks like Ireland was a blast, I’ve wanted to get over there for years. We’re seeing this with many client sites as well, but I’m not willing to give the developers as much credit as you with regard to trying to maximize Toolbar PR. I’ve had the pleasure of working with some devs recently who just don’t have a clue what URL structure is. They’re not trying to take advantage of a PR glitch, they’re just trying to do as little work as possible in the process of reselling some POS template site that they sell to unsuspecting clients.
Anyway, the same issue is the result. With hundreds and hundreds of files in the root we face the daunting task of adding structure if we want to stay sane, yet we are concerned with value evaporation with the implementation of 301s.
This might seem like a silly set of questions, but I’ll ask anyway. To try and avoid that evaporation (rankings loss), would you consider creating a duplicate content issue with the implementation of a solid URL structure while leaving the original page in place and then inserting a canonical tag on the original page page to solve the dupe issue?
Do you have a feel for whether the canonical tag evaporates more, less, equivalent juice than a 301?
Obviously, this would only be useful if the canonical method had less negative affect on rankings for a page.
Wow. A blast from the past! It’s kind of funny to see the different tactics that SEO’s tried back in the day. Seems like they would do anything to get ahead quickly, when it was really about taking a holistic & user-friendly approach all along.
IMO, the problem was that people mistakenly thought it was the URL and the lack of directories that gave them the PR the way they saw it. When in reality it has nothing to do with that and everything to do with how each page of the site was linked to each other.
You can have a URL that appears to be 10 directories deep, but if it’s linked in the main nav. to and from every other page of the site, it will likely have the same PR as one in the root linked the same way.
Deploying clean SEO is a beautiful thing.
While reading this article I was struck with the thought that SEOs and their clients could use a dose of copywriting courses to understand how to base their URL structure.
(Clean, crisp, and attention grabbing = Good … Keyword-Stuff Mumbo Jumbo = Bad)
Hope Ireland was a blast…sorry you had to refuse a beer for the first time in your life!
Great post, and I haven’t run into any site structures like that in a while. But I am sure that they are still around!
Jill had a great point regarding the internal linking structure in relation to page rank and what not. Totally true.
People often mistake a “flat architecture” with having ZERO directories. Maybe you should elaborate on that concept for everyone out there.
Thanks for a great article! Here’s a related page that always brings a smile on my face:
http://www.seoconsultants.com/s/e/o/c/o/n/s/u/l/t/a/n/t/s/d/i/r/e/c/t/o/r/y/
You ever just updated the entire site’s url structure all at once instead of in smaller chunks? If so, did you see any penalty?
I’ve only done an entire site all in one go one time, and it was on a decent sized site. Took a few weeks for everything to switch over in the index, but I didn’t see any negative consequences.
At the end of the day PR numbers don’t really matter. It might be a way to gauge but overall it really doesn’t make a difference.