Misdirection is something that stage magicians do to get you to look away while they slip your watch from under the handkerchief before hitting it with a hammer. It’s also something that you may be doing by accident with your site and, in the process, hurting your business.
You may think that you don’t have an issue, because you’re using the canonical tag, or you’ve got redirects set up on your pages to ensure that anyone who comes to your site (including the search engine spiders) ends up on the URL you want them to, with the content that you want them to see.
Here are five different issues that may be causing you problems.
Incorrect Use of the Canonical Tag
If you’ve implemented the rel=canonical tag to specify to the search engines which version of the URL you want them to use, regardless of the one that they actually reached the content using, then you’re most likely ahead of at least some of your competition. You’ve reduced the potential for duplicate content issues, and can feel free to use tracking parameters to your heart’s content.
Assuming the right URL is actually in the canonical tag.
Last year I recommended to a site that they use canonical tags. They asked for an example of a site that had implemented them well. I passed that site on to them, and their developers spoke to the developers of the other site.
They decided that the easiest thing to do was to cut and paste the code that generated the canonical tags. Unfortunately, this also meant that they copied and pasted the hard coded string that set the domain in the canonical tag. So all their canonical tags pointed the search engines to the other site, which 404’d and got indexed by Google (this happened about a year ago, and I’ve not seen canonical tags that 404 get in the index since, so Google may have fixed this shooting-yourself-in-the-foot-with-canonical-tags issue by simply ignoring them in cases such as this).
Do your pages have a trailing slash (i.e., do they look like www.example.com or www.example.com/)? If one redirects to the other, you need to make sure that your canonical tag contains the one that it’ll resolve to (if the non-trailing slash version redirects to the trailing slash version, then your canonical tag needs to have the trailing slash).
This should be a fairly easy one to spot, as none of your users will be able to get to the page as the page redirects to itself, or in some cases redirects to another page, which then redirects back to the original page, and so on. The browser will whir for a couple of moments, then inform them of the error. But if you have a large site, this may not be as obvious.
The best way to identify infinite redirects is to look in the crawl errors section of Google Webmaster Tools where they’ll tell you all of the issues they’ve encountered while attempting to crawl your site.
Your page was originally at URL A, then moved to URL B, so you put a redirect in place to send any traffic and link juice that was going to URL A to URL B. Then somewhere down the road the decision is made to move to URL C. So you put a redirect in place to send any traffic and link juice that was going to URL B to URL C.
That’s all you need to do right? No. You now need to go back and change the original redirects from URL A to URL B to instead point directly to URL C.
If it’s been a year or so, and you’ve managed to go back and change all of the old links from URL A to go directly to URL B, then it’s less of an issue, but it’s still a good practice to make sure that there’s only a single redirect where you can. One site I worked with changed domains twice in the space of four weeks, getting these redirects right was absolutely something that we had to do to ensure that we preserved as much of our traffic as possible.
Using 302 Redirects Rather Than 301 Redirects
A 302 redirect is a temporary redirect, and should be used when you anticipate the URL you’re redirecting from coming back at some point (i.e., for a seasonal service that’s only active for part of the year). A 302 redirect sends users and search engines to the new URL, but don’t pass the link juice or page equity to the new URL.
A 301 redirect sends everyone to the new URL and passes some of the link equity to the new URL. If you’re using 302’s from a page that’s never coming back, you’re not taking full advantage of what you’ve already built up for that page.
Linking to a Non-canonical Form of a URL
If the canonical form of your page is with a trailing slash, you 301 redirect the non-trailing slash version to the trailing slash version, and you have the trailing slash version in your canonical tag. Make sure that your navigation links all point to the canonical version.
Many sites don’t have the correct form of the URL in their navigation, footer links, and article links – all they’re doing is wasting some of that link juice. Make sure that your links on your own site use the canonical form of the pages they’re linking to.
Misdirection Correction Tools
Here are a couple of tools that can help you diagnose these issues.
- Microsoft IIS SEO Toolkit – This is my initial go-to tool for any new site that I look at. It crawls as many pages of your site that you’d like (up to 999,999) and reports on errors such as those listed above. The major issue with this tool is that you need to have Windows Vista or Windows 7 for it to work. It also obeys robots.txt, so won’t crawls a staging server that’s blocked to crawlers.
- Bruce Clay’s Server Response Checker – There are other server header response tools out there, but this one gives the results in a nice screen grab format. Simply put your URL in the tool, and you’ll see what redirections are in place for that page.