The duplicate content issue and the keyword cannibalization issue are very similar in their effects on your site’s traffic. There isn’t a duplicate content penalty and Google won’t ban you for either of them, but they both spread out SEO juice into multiple pages that fight each other for rankings for the same terms.
Identifying duplicate content problems is kind of like tracking an exsanguinating deer in winter; it’s not hard to identify the signs of stumbling, bloody fauna and you generally know what you’re going to do with the thing once you find it. In the routine duplicate content check, you know you’re going to have to fix multiple session IDs, redirect archived and re-purposed content and add (or correct the usage of) canonical tags, etc, etc… The keyword cannibal is a much subtler beast.
Grotesque hunting imagery aside, keyword cannibalization is an issue that affects many a webmaster, whether they realize it or not. And what’s more, even though the problem is almost ubiquitous, it is nearly impossible to control completely. If your site is even relatively focused with a modest amount of content, you’re going to find that somewhere you’ve got two pages ranking for the same terms. How can we control the problem, then?
Point #1: About half of the traffic that will reach your site’s pages will come from keywords that you did not optimize for.
First, let’s lay some groundwork. Some time ago, I started my first official website. I watched every day for that first visitor, my first search engine referral. A few weeks after launching my site and updating it, a few search referrals started to trickle in and it confused me; they weren’t searching for the keywords I targeted. The longer I watched, the more this trend presented itself. However, once my traffic stabilized, I saw a trend I didn’t anticipate; just under half of my search traffic came from keywords I had no intention of targeting. They were related to the page they landed on, sure, but I’d never dreamed I’d get the number of visitors I did for keywords I didn’t target. I’ve noticed this trend over many of the sites I’ve worked on.
On one site I worked with, keywords they targeted (for simplicity, those included in title tags and placed in content) represented 52.8% of all of the search traffic. This means that a full 47.2% of traffic came from untargeted keywords. But what does this all have to do with keyword cannibalism?
Point #2: Sites with very similar topics net similar long-tail results.
Best practices suggest that we should try to optimize a site for only a handful of keywords. This approach is a good one, since it’s nigh-unto-impossible to rank a single page for more than a few keywords, especially highly competitive ones. However, this approach presents problems in a site strategy when we give into the temptation to hone our focus into a laser pointer and forget the first point. By focusing so strictly on a small number of phrases, we neglect long-tail phrases and we’re tempted to use close-cousin keywords for additional content. On the previously mentioned site, they had a total of seven pages targeting the same type of product, but separated them out into individual pages by targeting slightly varied keyword phrases.
A cursory glance at some SERPs showed that for the exact phrase each page was optimized for (namely, the title), they each showed up in their own page and there appeared to be no problem with crossover. However, once I ventured outside of the handful of main keyword phrases, I saw a startling issue occurring with greater frequency: multiple pages (usually 3 or more) showed up for similar keywords. Sadly, they never hit the #1 spot for any of the phrases, but they frequently occupied spots 3-5. All of that wasted SEO juice!
Point #3: Multiple sites/subsites on one domain ranking for the same keyword is a waste of SEO juice.
You want those links and keywords and all the other tasty SEO goodness to be as focused as possible to boost you to the top of the SERP. I don’t think I need to go into how much more valuable a #1 ranking is than a #2 or any other rank that falls below that line.
So while on the surface, the tactic of creating multiple pages that target the exact same niche but with different keywords can net you a more dominant (but not necessarily #1) position for the two phrases you target, it can hurt your overall site performance by spreading traffic from long-tail phrases into multiple pages that are ranking for the same queries. Spreading the SEO juice love too thin nets lower overall traffic.
So what do we do about this?
Keep site keywords/topics different enough to avoid cannibalization. Keeping the targeted phrase on a page distinct and unique enough to generate its own juice and not step on the toes of other pages is difficult, especially if you deal with semantically similar topics. A good rule of thumb is that if a page’s targeted phrase shares two or more words with an existing page, you will likely cannibalize long-tail results. Just remember: you don’t create a site or page to rank for just one keyword.
So, if you’re thinking of creating a new page/post for a specific keyword phrase, give Google a cursory search for the phrase with personalization off and see if you already rank for the term (learn to make your own rank checker with Google Docs here). Then, if you already have a position for the phrase, see what you can do to boost that page for that term; build links with matching anchor text to the page, add some fresh content to the page/post about the phrase you’re targeting, and so on.
My pages are already cannibalizing, what can I do?
The solution for historical keyword cannibalization is the same as for duplicate content; toss some 301 redirects on the pages that rank lower to the top one and add canonical tags if the pages are duplicates.
What do you think? What can webmasters do to avoid cannibalizing long-tail keywords and solve existing cannibalization problems? Tell me in the comments.
