Real On Page SEO: A Content Cannibalization Debate Worth Talking About

[chanilla]

Let's talk about something worthwhile regarding onsite seo. By now you've read or contributed to the On Page SEO guide in Day 8, but what about your personal threshold for cannibalizing your own site?

We see evidence all around us on forums with multiple threads ranking with similar titles, yet constantly see the relevance shift and recommendations to take down multiple terms with fewer total urls. (i.e. - my interpretation of that is create better pages with more value)

Without making some fancy mind map, this MOZ snowboard example is so so enough to get the gist across, yet I would personally choose to integrate them all in one. In some cases, I see where this is not practical since you'll blow your kw density out of the water in cases where all product prefixes are the same, and are constantly referenced to create context.

So let's say AHREFS wanted to enter the web hosting review space, and expand their reach into WhoIsHostingThis' britches. They fire up the scrapers and start compiling low hanging fruit with purchase decision intent for their "writers" to get busy on. We'll assume they're sticking with ultra long tails, since the their ability to create useful "data-backed" content is akin to the usefulness of a bag of dicks to a pornstar.

25 Cornerstone Pages:

• Best Web Hosts
  • For Bloggers
  • For Businesses
  • For WordPress
  • For Joomla
  • For Dick Pics

• Hosting Comparison: The Ultimate Guide
  • Hostgator VS Fatcow
  • Hostgator VS Bluehost
  • Hostgator VS Godaddy
  • Hostagtor VS ASmallOrange (FFS thank god you guys stopped promoting this company!)
  • InMotion VS Hostgator

500 Blog Posts: (mostly Q&A style, since personal research data is hard to come by)

• Which is the best host that supports hot linking Tim's dick pics?
• What's the best host for a personal website?
• The Top 3 Web Hosts That Will Boost Your SEO Like AHREFS Blog!
  
• Where's the best place to host my blog?
• Which is the cheapest most reliable web host?
• etc...

We know that urls don't matter, links maybe a little bit, and on page... Meh why bother. /ahref training sarcasm

Shitty overall site architecture example, but you get the idea. In the past I would have done the above, but am noticing more and more (pertaining to the above example) creating Best Of FAQ posts with multiple questions answered in one go, tends to perform better. I would attribute this to length, number of long tails integrated into earlier page age, more links both external and internal (less places to link to, higher the frequency), etc...

I'm not sure where the threshold for cannibalizing one's own site is anymore, regardless of all the canonical bullshit or not. Forget about it and publish, publish, publish... Or be cautious and account for future animals with big teeth.

Threshold of content attrition/cannibalization...

Myth or fact? What's your "data-backed" take on this pal?


If you find this to be useless, continue the discussion with something useful below...

 

[Kevin]

I have been trending towards the latter method you described, going after keywords in clusters that add up to around 2x the volume of the main keyword I'd like to rank for. Answering more, covering more, and going deeper in single posts. We'll see how it goes.

 

[Ryuzaki]

The goal in busting your giant article into tons of small ones linking to a main hub is to rank for all of the terms across many pages. (It also helps with tighter on-page and less over-optimization).

But at the same time, the goal of keeping it all on one giant page is to rank for all of the terms on one page.

I personally go for the 2nd option. I want it all and I want it on one page. I believe data supports that this is the best approach currently. SerpWoo, SEMRush, and another other SaaS that can show you the terms that a page is ranking for will show you... when a page dominates, it really dominates. These are the kind of pages that lock into spot #1 and don't bounce.

Why this happens likely has to do with the completion of the topic. It discusses everything you need to know in a clean hierarchy. It's worth it for the big keyword and still supplies the answers for all the longer tails.

This level of completion implies length, quality, pics, lists, videos. Now we're talking about a lack of pogo-sticking back to the SERPs. Now we're talking about low bounce rates (if a page rocks, I'm going to check out the rest of the site). Long time on site, long time on page, views per user, etc.

Also, if you're trying to rank 10 pages for 10 terms, you're going to need to attract a lot of links, no matter how you interlink. But if you're trying to rank 1 page for 10 terms, life is about to be a lot easier. However many links it takes to get ranked for the big term is all you need to worry about. You'll rank for the long tails along the way.

Obtaining these links is going to be a lot easier because you'll have an actual linkable asset instead of MFA style content designed for robots instead of users.
_______________

Of course it's not that clear cut. The value of the long-tails and the competition of all of the terms may cause me to take a "busted up" approach.

Usually I'll get the best of both worlds. I'll write the giant article and then still build supporting pages that aim to rank for tangential terms.

With snowboards, I'd might try to rank for:

• Snowboards
• Professional Snowboards
• Speed Snowboards
• Beginner Snowboards

and then turn around and post some stuff for terms like:

• Snowboard Fail Videos
• Snowboard Accidents
• Snowboard Tricks

These are just as capable of linking back to the main page and boosting it as any of the other long tails would have been. Now I get to have one super page still.

 

[turbin3]

Trust me, most of you guys have probably not seen the depths of content/keyword cannibalization that I have. I'm talking a network that sees ~2BN hits/mth, consisting of several thousand separate domains, consisting of billions of pages, much of it thin, duplicated, or downright redundant much more than I would like...

Ryuzaki is on the money. Where possible, consolidate down, because it is simply much easier to manage and focus on fewer things. As time and resources allow, expand up to whatever degree makes sense.

I would go about choosing which to focus on first based on your time and resource constraints, as well as based upon the level of competition in the niche(s) you're targeting. Too tough? Go longtail until you're getting good traction, then shift focus to the higher level. Already on top? Come in for the kill with all that longtail and destroy hearts and minds.

 

[Samwise89]

This spray and pray crap is what Google doesn't want to have in it's index.

Not to mention... you don't need to rank for a million keywords. There are sites earning enough for people to retire off of from two or three rankings. Let alone a million.

Posting 1,000,000 articles takes forever, costs a ton, and you don't have time to market any of it. The only way to build any links is to spam, and that's self-sabotage.

Or you can write one post and promote the hell out of it and actually secure a ranking that's worth some money.

Focusing on a million things isn't focus.