Way back on Feb 12, 2009, Google Webmaster Central blog added a post titled, “Specify your canonical.” (Yes, title case seemed to escape Google at the time.) “Carpe diem on any duplicate content worries…” they told us.
And e-commerce webmasters breathed a sigh of relief. Finally, they could run duplicate descriptions for their merchandise on their front page, the individual product pages and wherever else they wanted to as long as they specified ONE page as the canonical or “preferred” page.
Cool. Prior to rel="canonical" e-commerce stores had been running amok trying to create different titles and descriptions on every page where an item was sold. Imagine the stress level at an e-commerce store the size of Amazon, right? Problem solved. Thank you, Google.
But other webmasters, who were eager to share duplicate content all over the Internet found that canonical didn’t mean much for them. But let’s back up…
What Does the Canonical Link Look Like?
This tag goes into the <head> section of your page:
<link rel="canonical" href="http://www.page-title.com/" />
Simple enough. Google takes this “hint,” which they say they “honor strongly,” and understands the preferred version of a page. It’s OK to have exactly duplicate or mostly duplicate, too.
So, if you have category pages, affiliate pages or any other pages with identical content and you add the canonical tag showing the preferred page for each of them, Google understands and won’t smack you with the dupe content on your own site penalty. That smack can happen swiftly and hard. Just don’t do it.
But here are some things you may want to know about:
- Canonical URL properties like PageRank and other positive factors are transferred to the duplicate pages, as well.
- Seems like a no-brainer, but Google won’t recognize the canonical, if they haven’t yet indexed the page.
But there was little relief for webmasters who had duplicate content on multiple sites. You could not use canonicals across domains.
Canonicals Get Smarter
In December 2009, Google announced it would then support Cross-Domain rel=”canonical” link elements. Again with the, “Yay! I can post articles on 10,000 sites on the Web!”
Well, you can, if you want to tag each version with a canonical tag that points to one preferred version of your content, and hope that Google agrees with you. Yet, there’s really so much that could go wrong, I’d advise against doing it — strongly. Something breaks and you’re bitten by Google’s big Panda. Nobody wants that.
Plus, that’s not why cross-domain canonicals were instituted. They came about because people were moving sites to new domains, and some servers won’t allow 301 (permanent) redirects. Therefore, http://OldDomain.com became http://NewDomain.com, and all of the OldDomain.com pages became broken links.
People were losing the influence they’d built on the Old Domain, but they couldn’t let spiders know, “Pssst… We moved. Here’s our new page.” So, by using cross-domain rel=”canonical” these webmaster could tell spiders, “Hey look guys, we know that this content is identical to the old site, but this new version is the preferred version now.”
It’s like changing your address at the post office, right? If you want your mail…
Just remember that a 301 redirect is always the preferred way to handle a moving situation, but when you have no other choice, cross-domain canonical tags can and do work.
What Could Possibly Go Wrong?
OK, so Google gives us cream and we make buttermilk. Or something. We start using canonicals for the wrong reasons, and now, Google needs to remind us of a few important things, IF we want our tagging to work properly.
In April of 2013, they told us we weren’t doing things right and gave us these hints:
- Make sure that a large portion of the text on the canonical page matches the duplicate page. There can be small variations, but overall, they should seem very similar. Dr. Pete over at Moz wrote he has tested this theory and found that really dissimilar pages could rank using the canonical in some cases. But you probably don’t want to take the chance.
- If you’re going to use the canonical be sure that the preferred page exists. That’s another no-brainer, but let’s face it, people make mistakes when typing. Or, you might be using a canonical that no longer exists on the Web. If you own both pages: when you take the old page down, fix the canonical tag on the new page and make it the preferred version.
- Never use noindex on the canonical page, either in a robots tag or robots.txt. Since the canonical page is supposed to be the preferred, so this is a fairly silly mistake.
- Be sure that the rel=canonical tag lives in the <head> section of your page. If it’s in the <body>, the tag will be ignored.
- Never specify more than one canonical page for the same content. If you appoint more than one, guess what? They’re all ignored, meaning you have dupe content. Argh. Not good.
- If you have an article that spans three pages, don’t use canonical links from pages two-on to point to page one. Otherwise, the 2nd and so on pages won’t be indexed. Instead, use rel=”prev” and rel=”next.”
- Make sure when copying & pasting identical page styles with varied content that you don’t leave a rel=”canonical” tag in place for the old page on the new one. Otherwise, the new page will never be indexed.
The Bottom Line
The bottom line is this: use the canonical link. It’s useful if you have duplicate content on your own site or on multiple sites. Just be sure that it’s in the right place, that you’ve typed your tag properly and used it wisely.
Otherwise, you’ll just be wasting your time, causing you a big, fat headache. Not me!