The definition of cannibalisation, in SEO terms, is when there is more than one page being optimised for a particular search term, causing confusion for search engines. They can’t decide what to show in the results page and therefore display different URLs from the same site. This is known as cannibalisation because the site is seen as ‘eating’ itself, i.e. competing with its own pages, for the same search term.

Whether it is an article, similar product pages for the same item or similar URLs with similar words, they are all seen as duplicate content and become an issue for indexing in Google, Bing, Yahoo, etc. There are four types of cannibalisation that can occur on a site; internal, sub-domain conflict, international conflict and semantic flux, not all of them have to appear on your site for there to be an issue either.

Internal Cannibalisation

This is when your site has more than one page optimised for a particular search term and Google can’t decide which one to show in search results pages, so displays more than one. This has a negative effect on your site and its ability to rank for your chosen topics, meaning your site will not reach page one consistently and your position in Google results will always fluctuate.

Sub-domain Cannibalisation

Sub-domains can open up a lot of issues for many sites, cannibalisation and duplicate content are just one. If your site has a featured sub-domain; a blog or online store, and you optimise for the same search terms on both your main site and the sub-domain, you will cause confusion in search results, once again. Duplicate content will cause both your sub-domain and your main website to appear in results pages for the same terms, meaning you have sub-domain cannibalisation. Once again, this will mean your results and ranking will never be consistent, there will be conflict of interest from a search engine's point of view and it will render your sub-domain as self-damaging to your site.

International Cannibalisation

This happens to larger brands and websites that need to have multiple versions in different countries, emphasising how intelligent search engine bots can be when it comes to identifying a common relationships between web properties and companies. Coca Cola are a great example of how larger companies who need multiple websites for each country, avoid international cannibalisation. By treating each of their web properties eg. .co.uk, .com and the .ca sites, as unique and individual entities, despite featuring similar products. Their content is very different for each site and everything from layout to featured products are different, ensuring search engines never confuses a site.

Semantic Flux

This tends to occur for larger companies and corporations who have sub-brands or smaller companies within their main corporation, selling the exact same products or featuring the same promotions across brands. Semantic flux is when similar themes feature across sub-domains or other sites that are semantically related (via brand or corporation) even if their content is generalised. Search Engines can identify the relationship between brands and companies from historical data and will confuse results by showing both sites for the same term. If there is no clear canonical page on either site for that particular search term, the search engine will simply pick both. This again will mean the sites will always struggle to remain in page one for results and their results will continue to fluctuate.

For a more in-depth approach to the types of cannibalisation and how to deal with them, as well as some great examples of large companies getting it wrong, we recommend watching Jon Earnshaw’s speech at Brighton SEO, this year. Jon explains in great detail about how cannibalisation can be created and what you can do to prevent it.

How to deal with cannibalisation

The first rule of handling cannibalisation of any type, is to address and check internal cannibalisation first, as this is more important and fundamental to addressing any type of duplicate content. If you can identify internal cannibalisation within your site, there are several actions you can take to resolve the issue of duplicate content. Begin by choosing just one page to optimise for one particular search term. Conducting extensive keyword research is highly recommended and if you have someone working on your online marketing, this will be something they need to conduct. Remove the specific term from the page title or URL and strengthen the internal linking within the page content, ensuring you link back to the original and most relevant page. This will clear up any confusion search engines would then have, ensuring your site only appears once in the results page.

Dealing with sub-domain cannibalisation is very similar to dealing with internal cannibalisation, in that you need to decide which site should be optimised for particular search terms. Focus on just one of them appearing for particular search terms and remove any similar or exact-match content and phrases, applying 301 re-directs won’t be the answer in this instance unfortunately. The other option is to simply remove your sub-domain! Don’t use them unless absolutely necessary, and if you choose to work with one, make sure you know how to handle it correctly as they can create far more trouble for your site.

Another good resolution for this problem is implementing 301 re-directs. 301 redirects will become your new favourite, as they simply direct traffic away from the duplicated URL and back to the original canonicalised page. Ask your developer to implement these as they are permanent and if executed incorrectly, they could do more harm than good. By redirecting traffic away from the duplicated content, you are telling search engines to ignore one page and focus on another. Optimising just one page for a particular search term will mean your position in search results will be more consistent and stable, with little fluctuation and hopefully far more traffic!

Using a rel=canonical tag across your site is a must if your site suffers from duplicate content as it informs search engine crawler bots to see each page as one piece of unique content. A rel=canonical tag is a small piece of code which sits in the header tag on your website, informing search engines to treat each page as individual content. A rel=canonical tag will compliment your content changes further once your content is in order, treating each page as unique sources of content. Ask your web developer to add this small code to your site or check out Moz’s run through of rel=canonical tag implementation.

Noindex follow is another tactic for telling search engines not to index particular pages on your website, within search results. This is particularly useful for much larger websites with thousands of similar or duplicate themed content and is also slightly more tricky. This should only be implemented by someone who knows their way round webmaster tools and is confident in de-indexing pages.

There is new argument for a 5th type of cannibalisation emerging, as social sites begin to dominate the SERPS. Social button cannibalisation is when your social channels either outrank your actual website, or they are seen creeping into 2nd or 3rd positions below, causing visitors to opt for their shiny, social buttons, by-passing your website and leading them straight to social pages.

No matter what type of website cannibalisation is caused, it is very important to address it if it is happening on your site. There are many ways to prevent duplicate content from creating this issue, from careful keyword research and placing across your site, creating unique and interesting blogs or articles and never copying and pasting your copy. If you are guilty of copying and pasting page titles or page content, re-visit those pages and ensure they don’t have the same content twice! Although search engines are simple robots, they are intuitive enough to identify a subject or word being repeatedly mention on a website. Make sure you always generate unique and original content for your site.