How serious is keyword cannibalization?

[hiruma]

How serious is keyword cannibalization? Would it destroy my site forever or something?
I'm currently doing the avalanche technique and the topics I'm writing are so closely related that I sometimes feel like I'm just repeating myself. Though I checked the keywords and I know that they have a slightly different intent.

I currently have 50 posts and I'm writing 3 posts per day. I'm worried that after a few months I wouldn't be able to track if the keywords I've written before having similar intent to the ones I'm writing.

 

[Ryuzaki]

How serious is keyword cannibalization?

Serious enough to affect your ability to maintain a consistent and high ranking for keywords that are being cannibalized.

Would it destroy my site forever or something?

No. What happens is all of the pages going after the same keywords (accidentally of course) can compete against each other. Google won't know which is the one that should be ranking so they'll keep testing them all (and your rankings will bounce). Often, all of the rankings related to this problem will rank lower than they should for whatever reason. It's almost like "We don't know which is which so we'll demote both of them a bit".

Though I checked the keywords and I know that they have a slightly different intent.

Separate intent definitely helps with this problem, but sometimes you're going to run into it regardless. Especially eCommerce sites with product variations. What I'd recommend is to try to rank category pages most of the time and use faceted navigation to let users narrow down, and not try to rank for every minor variation of the main keyword.

On content sites you run into problems when you try to rank for keywords with separate articles like:

• Best Widgets
• Best Widgets Under $100
• Best Widgets for Kids
• Best Widgets for Beginners
• Ultimate Guide to the Best Widgets

And so forth. If the intents are clearly different like they are above, you can minimize the problem. You really never know when Google is going to segment a SERP with multiple intents in it (so suddenly you lose your "best widgets" ranking for the Best Widgets page and it's replaced with your Ultimate Guide page, etc.

It's really just something you're going to deal with in some cases. Google will likely get better at dealing with it over the coming years. But you should be avoiding it when possible. Stuff like:

• How to Click my Mouse Button
• How to Click my Mouse Button When it's on my Desk
• How to Click my Mouse Button With My Finger

Those kind of variations are going to cannibalize each other, especially when there's not enough volume and user feedback to help Google understand the different intents beyond natural language processing, or enough articles on the topic to rank.

 

[Philip J. Fry]

Stop worry about stuff that you're ignorant about.

You first off, you need to check if the two (or whatever) keywords are treated as synonyms or not. You can check by opening two browsers and googling one keyword in one and the other keyword in the other. Look at the results. Are the results very similar? If so, Google's determining that they are very closely related. If they are not, they are not related.

If they are treated as synonyms, don't write another blog post about that topic.

While two keywords might seem like synonyms to you, that's not how Google's analysis of the Internet might conclude. Don't delete posts off of your site just because you have anxiety. Good bye.

 

[eliquid]

Depending how long term you want to keep your site and how serious you are:

1. I would write them anyways. As Google gets better, those pages might start to uncannibalize themselves. This is a "might" though.

2. Write a version for each page on another site, like a parasite website. A shortened version that you can bang out quickly. See what happens to that parasite over the next few months and how Google treats it. Based on what you see, adjust and see if you need to put it on your main site.

This action of course is dependent on how long you want to keep your site and how much time/effort you also want to put into it.

If these are low volume terms with little revenue potential, then I would not do this.

However, if we are talking high volume or high revenue potential.. then it could be worth it to you to invest the time and money to find out what will work best.

 

[bernard]

In my experience it mostly a problem in search terms with mixed intent.

If for a search you have various types of websites: informational, commercial, test etc, then having more than one post for a keyword might make you jump around a lot, as Google has to constantly decide what intent each of your posts has and how it fits in the search intent.

The more competitive, the more on point and the less confusion, is needed.

 

[TheCurator]

I'm currently doing the avalanche technique and the topics I'm writing are so closely related that I sometimes feel like I'm just repeating myself.

Idk all the ins and outs of this technique but if you mean that you are writing individual posts for very similar keywords that get low search volume please stop doing this. Gather the keywords into a group and use them in the same article as long as they have similar intent.

If these were your keywords...

• How many blue widgets can I fit in my car? - 0 searches
• How many blue widgets can a car hold? - 10 searches
• Blue widget capacity in cars - 80 searches
• Blue widgets in cars - 1000 searches

... would you be writing 4 articles? Please say no.

 

[hiruma]

Idk all the ins and outs of this technique but if you mean that you are writing individual posts for very similar keywords that get low search volume please stop doing this. Gather the keywords into a group and use them in the same article as long as they have similar intent.

If these were your keywords...

• How many blue widgets can I fit in my car? - 0 searches
• How many blue widgets can a car hold? - 10 searches
• Blue widget capacity in cars - 80 searches
• Blue widgets in cars - 1000 searches

... would you be writing 4 articles? Please say no.

I was actually kinda doing something similar to this. But the difference is I would write a big article targeting the keyword Blue widgets in cars that includes How many blue widgets can I fit in my car?, How many blue widgets can a car hold?, Blue widget capacity in cars as H2.

Then, I would write separate articles for How many blue widgets can I fit in my car, How many blue widgets can a car hold, Blue widget capacity in cars keywords if the keyword needs more detailed articles because I don't want the main article to be 4000+ words long.

Correct me if I'm wrong but from what I understand from the avalanche technique is that the supporting articles need to link up to the main article. Kinda like a pyramid where the low-volume keywords support the high word count article.

 

[Ryuzaki]

Correct me if I'm wrong but from what I understand from the avalanche technique is that the supporting articles need to link up to the main article. Kinda like a pyramid where the low-volume keywords support the high word count article.

You're not wrong, but any Keyword Golden Ratio keyword, which is what the Avalanche Method is suggesting you go after, is not a main keyword. Anything you find that's low volume and low competition will not be for a main article, it will be for a supporting article.

Let's say this site is about Blue Widgets and the holy grail keyword is "Best Blue Widgets" and it's 5,000 search volume and a keyword difficulty of 30/100. All of your rinky dink "how many blue widgets can fit in my car" articles will link to "Best Blue Widgets" as the main article. There will never be a case where a 20 volume keyword at 1/100 competition level will be a main article, unless you don't plan on making any money.

because I don't want the main article to be 4000+ words long.

Why not? Your alternative is keyword cannibalization on the ones in this specific example, and that can be generalized for most cases, really, including my example about for kids, for teens, under $100, under $500, for beginners, etc. You don't want to include those in the main article any more unless you want problems.

 

[tomer]

I was actually kinda doing something similar to this. But the difference is I would write a big article targeting the keyword Blue widgets in cars that includes How many blue widgets can I fit in my car?, How many blue widgets can a car hold?, Blue widget capacity in cars as H2.

Then, I would write separate articles for How many blue widgets can I fit in my car, How many blue widgets can a car hold, Blue widget capacity in cars keywords if the keyword needs more detailed articles because I don't want the main article to be 4000+ words long.

Correct me if I'm wrong but from what I understand from the avalanche technique is that the supporting articles need to link up to the main article. Kinda like a pyramid where the low-volume keywords support the high word count article.

As far as I am concerned, this is a terrible approach.

I can understand writing a big article which includes all the different variations of the keyword. In fact, I think this is good SEO practice, as not only should it help you rank for the different variations of the keyword, but I think it should also help you rank for the main keyword (by demonstrating relevance).

(Even then, I'm not sure whether I wouldn't combine How many blue widgets can I fit in my car? with How many blue widgets can a car hold?)

But writing separate articles for each of those variations?! At least with the example keywords provided here, this sounds like textbook keyword cannabilization to me.

I like to look at it from the perspective of user intent, and in my opinion, all of those articles address the same user intent. So you're basically creating duplicate content.

I think a better approach would be to write an overarching article about blue widgets in cars, using some of the keyword variations you mentions, and write separate, supporting articles about slightly more specific variations of the keywords, where the content might actually vary.

For example: How many blue widgets can you fit in a saloon/coupe/estate/sports car/etc?

Or: How many blue widgets can you fit in an Audi/BMW/Mercedes?

I think I'd personally go with the former. And by the way, you could cover all those question in the main article too, and I think you could still have keyword cannabilization issues. But at least the articles focus on a slightly more specific intent (instead of exactly the same intent in different words).