Is this keyword cannibalization?

[PaulEchere]

Hi,

A little while ago I heard about creating "supporting" pages to drive more organic traffic improve positions, etc.

For example you have a page "Some casino / sportsbook review" - the page is a pretty detailed review of the brand and one of the sections of your review is their mobile app. So you go an create an entire page devoted to that brand's casino app and link to the new app page from the review page.

Does such an approach make sense or would you rather consolidate both pages on the review page?

I've taken a look at several well performing affiliate websites and there doesn't seem to be a single approach to this. In some cases review (or any other similar large pages) branch out into smaller and more specific ones and in other cases everything is kept on a single page.

Does anyone have any experience with trying this and seeing how it performed before and after?

[universal4]

I think that you have a logical point but I think the reason there is not a single approach is that it can happen more naturally.

I think that over time as a person adds articles instead of adding "casino1review #2" that would contain a lot of the same content it makes sense to write articles that cover certain aspects we'll call them "casino1aspect1" and "casino1aspect2" then linking them to the internal page "casino1review" makes very logical sense for both site structure as well as for the visitor.

To visit the casino "click affiliate link" to read more "casino1review".

I think this just goes with the theory of getting the visitors that are taking the time to read the content to read more until the right article closes the sale.

I think in this scenario it is important to have clear calls to action on each page.

Will be interesting to see how this discussion develops.

Rick
Universal4

[chaumi]

Complicated aspects to this with many nuances.

But, using an example or two, I'll have a go...

So, very recently, at least one (and I think two) new pages went into the SERPs fairly high up for 'bet365 review'. They're not 'newspaper' pages as such (as in The Guardian etc etc etc)...but they are very well-known (at least in the UK/Ireland) sports betting news, odds, media, betting guides etc etc sites. So they will have authority around sports betting in general.

We'll take one of them (I won't link for obvious reasons, not least because they don't deserve it!)

So, they've put up one of the weakest (ranking) pages on bet365 I've seen in some time, on a par with all the casino/bingo etc pages we all know that shouldn't be there. But it is well written, though. Grammatically correct, reads well, appropriate terminology throughout but pretty basic.


Here's what they rank for within the space of a few weeks since going live:

bet 365 horse racing
bet365 racing
bet 365 horse
bet365 horse racing best odds
horse racing bet365
bet365 horse racing odds
365 horse results
bet365 horse racing results
bet365 odds
grand national odds bet365
bet365 tv
bet365 irish grand national
bet365 best odds guaranteed
bet365 best odds
grand national betting bet365
bet365 grand national matched betting
bet365 withdrawals
best odds guaranteed bet365
bet365 best odds guaranteed time
grand national bet365
how many places do bet365 pay grand national
bet365 withdrawal limit
bet365 national offer
bet365 paypal
betr365
bet365 fast withdrawal
bet365 minimum deposit
bet 365 free bet
bet365 new account offer
free bet365 bet
bet365 mobile site uk

The top 9 or 10 on page 1, 10-15 on page 2/3, then progressively down into the 60s, 70s etc

No obvious backlinks of any note, so appears to be ranking on pure site authority/page content alone (and maybe just freshness, but we won't know that for a while).


Let's start...

Firstly, this is a Bet365 review. It was found by a search for 'bet365 review'.

The first question is 'what did I want when I searched bet365 review?'

Simple question, but not so simple answer, right?

Bet365 offers sports betting, poker, casino, fantasy, bingo, esports, etc

What did I want?

I know what I wanted. A review of bet365. High level. All of it.

But how does the algorithm know that? In the absence of any other signals, it doesn't. But it does know that people search 'bet365 review' for different reasons. Casino players want casino, sports punters want the sportsbook, bingo players want bingo, etc. Not all searchers know to be precise. Sure, there are elements of a bet365 review that are generic, but when you get to the detail, some elements will/may be significantly different.

So, if someone doesn't specifically state what they want (ie, for example, bet365 casino review),it will show a mixture/a selection of results giving it the best chance of somewhere answering the query. Indeed, it may also show results showing different aspects of the more precise 'casino review query, for the sake of offering varied options to searchers.

That (I believe) is logically why this page, which is specifically about Bet365 horse racing, ranks for the high-level term 'bet365 review'. (ofc, it could be just the algorithm is screwed up, but let's give it some credit and accept we may be wrong in doing so).


************

Bear with me, this is all relevant.

*************

So, look at the terms. Look at the variety. And you could pick out n number of them that you (or they) could build as supplemental pages.

Let's take 'Bet365 Grand National betting' as an example.

So, we build a page named /bet365/grand-national-2023-betting/

And fill it with everything related to bet365 grand national betting - odds, fancied horses, path to the national, who's targeting the race, any horses that have already performed well in 'trials' over the National fences, etc etc etc etc etc etc

Our target is to rank for bet365 Grand National betting.

***********

In theory, we've got a potential problem. We already rank for '365 grand national betting' with the review page. So, there's a chance our new page might be seen as involved in some sort of cannibalization scenario.

Obviously, the hope is our new dedicated page ranks for the terms..but...if it doesn't..and we really want it to...then we have to (try to) eliminate the reason why the original page was ranking for it (or try something else, as we'll see in a minute).

Note : in this case, the original page mentions the Grand National very lightly, so just removing that reference might be enough, but hopefully you'll recognize it might be significantly more complicated than that in other scenarios.

Or...we don't eliminate it...and rely on other signals to tell the algorithm our new page is the resource they want to classify as more authoritative - in other words: backlinks, and appropriate anchor text, of course. An approach that may or may not work.

***********

So, why go into this detail?

It illustrates the problems you face in trying to second-guess cannibalization from the outset.

With the SERPs all over the place and often trying to answer queries with different (but not in some cases obviously different) intents...unless you are doing something which clearly and obviously will likely result in a cannibalization problem...then the most sensible way to handle it is to identify (with careful thought) the supplemental pages you want to cover, build them (sensibly/effectively)..and go for it.

You wouldn't want to lose rankings first for another page anyway, so be prepared that you may either have to make amendments and tweaks to fix it if some sort of clash comes up and/or send other identifying signals.

***********

So, applying it to another example.

You have a page on Nigeria sports betting. High level, about sports betting in Nigeria.

You have another page on Best Nigeria sports betting sites. That's a legitimate page to have as 'supplemental' to your main page.

But if your best sites page already appears in the SERPs for 'Nigerial sports betting' (which it might, and indeed, I did find one easy enough) then you may have a cannibalization issue. Even though it's a legitimate page to have.

Where you could sensibly argue it's not legitimate - and probably a questionable approach to take - would be by having a 'best Nigerian sports betting' page and then building a '5 great sites for Nigerian sports bettors' page/post. That's asking for trouble. They're semantically too close.

But having your top-level page and then 'betting on tennis Nigeria' ' betting on football in Nigeria' etc etc would be a legitimate approach. There might still be cannibalization issues, but only if you were erroneously ranking for one of those terms under the main page (from a Google decision that they needed that result in the SERPs for the sake of alternative responses to queries, and/or you included something on that page that might have given them some indication your page might be a good fit for the query).


******

In summary, much of it is down to common sense, and recognizing/understanding how the algorithm sees and classifies your site and the content on it.

But, in many cases, if you apply common sense and structure/present your pages sensibly, you'll have done much of what you can do to prevent (or minimize) cannibalization. Then - where it does happen - it'll usually have a cause that is outside of your control.

*******

Secondary to this, but related, is the fact that if you keep adding supplemental info to long pages, you're potentially (and in reality) doing a couple of things...1. Increasing your DOM/html size potentially too much that it introduces crawling errors or limitations, and 2. just increasing it to a size that Google thinks to itself 'hey, no one's ever going to read all this anyway, I'm not going to show it, and I'll show one of these other 10,000 more succinct - but still accurate - pages instead.'


*******

I think all this really answers the question. Where appropriate, it makes a lot of sense - both from a structural, crawl-render-label/store/index-serve, and user satisfaction perspective, to think carefully about how you best present a higher-level topic/subject and then deliver supplemental info around it if required.

If cannibalization occurs, deal with it then. But don't avoid doing it because of that fear, just take as sensible steps as possible to prevent or minimize it. Some of those steps will be around how you target/name, some around how you populate the page(s), some around how you use anchor text signals, and some may be outside your control anyway (whatever you do).

[PaulEchere]

Thanks for the input. I've already started creating a number of additional pages like this so I will monitor it over the next several months to see what effect (if any) it will have on the "main" pages and how well will the new one perform.
Will probably share my findings (to a certain extent) here once there will be enough of them to draw a conclusion.

[PaulEchere]

To give a small update here:

In early December I published a number of supplimental pages, I am still monitoring what effect that will have, but checking things today I noticed something interesting.

#1 One of my better performing pages clearly suffered right at the same time when the "supporting" page started getting impressions. It suffered in impression volume and positions (less in position). Of course, this is only a correlation and might have had other factors involved.

#2 Other pages that didn't perform as well - didn't show any negative side effects.

For now I have consolidates both pages I mentioned in point #1 and will see how and if it will recover.

[chaumi]

"One of my better performing pages clearly suffered right at the same time when the "supporting" page started getting impressions. It suffered in impression volume and positions (less in position)."

Were (some of) the impressions for the supplemental page for the same keywords as the main page?

[PaulEchere]

That's a good question, which I failed to ask myself earlier.
Not really, the overlap of KWs is about 2% and the ones that I didn't really care about on the main page.
Now thinking about it - it doesn't necessarily explain why the main page started doing worse.

P.S. even though the in light of the above this correlation is less likely to be the cause, I will still give it some time to see if what I did will have any positive effect. 301ed the new page and merged the content.

[Oliver Cooper]

I would be interested to know the result after

[bpmee]

Just to throw in a couple exogenous considerations:

1) Site authority matters - High authority sites can more easily rank 2+ pages for related topics (Nigeria Sports Betting, 5 Top Nigeria Sports Betting Sites). I suspect partly because of crawl priority.

If a small site has 2 somewhat-similar pages, it's not worth their time to rank both. In other words, you can't "specialize" until your site is big enough to get a closer inspection. Exception might be if those pages have good links from different sites.

We see this with major media sites all the time. One day, they write about the stock market falling as news. The next, another writer explains 5 reasons why the stock market falls. Both have some overlap, yet both mop up a wide swath of stock market news and basic reference keywords.

On the other hand, a fledgling site would likely find Google prioritizing the first, largest page and ignoring the other.

2) Inbound links matter - Related to #1, but similar pages with different inbounds from other high-authority sites will likely rank independently.

3) Internal linking - Are you use the same anchors to link both pages or something different? If you treat the pages the same, why shouldn't Google? (I struggle with this one myself -- it's the 19th item on a mental checklist and sometimes life is busy)

4) Recent Helpful Content and Link Spam Updates - I don't know if this combo has officially ended yet. But, I do see some changes in the SERPs and occasional flux. Some sites got big promotions (deserved or undeserved) and others got dinged (also deserved and undeserved). The final stages of these updates could be messing up what would otherwise be a "normal" content review?

Chaumi's content analysis is pretty spot-on here. I think there are other factors besides pure content at play.

[PaulEchere]

I am still waiting to see what effect consolidating both pages would have.
But I did actually pay quite close attention to make sure that link anchors would be different, external and internal ones.

Also, as I discovered earlier the two pages had almost no overlap in KWs. None in the ones I would care about for sure.

But the dip for the "main" page strongly correlated with the introduction of the new one.
I have been doing some research and also found that one extremely authorities aff website got into rankings in that region and for those KWs, also around the same time.

[chaumi]

But the dip for the "main" page strongly correlated with the introduction of the new one.
I have been doing some research and also found that one extremely authorities aff website got into rankings in that region and for those KWs, also around the same time.

An illustration of the problems involved in diagnosing why something happened. In any normal day/period, there are multiple potential causes of a fluctuation/movement. You're looking at a period from Dec into Jan when there were (extreme, in some cases) fluctuations due to updates and (possibly) whatever Google do every year at Christmas to bring in more pennies. In addition to 'normal' causes like new entrants, recognition of links, etc etc etc

[chaumi]

bpmee Multiple bonus points for the use of the word 'exogenous' (in addition to the highly perceptive sentiment overall). I like it, stored for future use!

[Lasmanis24]

Its depends of country where you target kw. Sometime is better make in one page this, sometimes need separate pages.

[gm2891]

Have look at this site: bestcasinosites. net, and decide whether you need it. If you're on your own, then I don't think so, but if you have some 50 spare hands, then yeah give it a go. They have 5 pages in total for some of their casino reviews (bonus, live, mobile, terms, and an actual review page).

[JohnLucky7]

Yeah, i guess you need to create a separate page !

[BestBTC]

Have look at this site: bestcasinosites. net, and decide whether you need it. If you're on your own, then I don't think so, but if you have some 50 spare hands, then yeah give it a go. They have 5 pages in total for some of their casino reviews (bonus, live, mobile, terms, and an actual review page).

That's a good site, super on point example, thanks!

[LinuxGam]

If google doesn’t give your site much trust/authority it wont make much difference.

I you have some trust/authority creating more pages on the same/brand subject as long ass its different and valuable information will increase the chance of the primary page ranking higher, particularly if you link back naturally to it in context.

[PaulEchere]

In case this might be helpful to some of you:

So I decided to consolidate the two pages - the main page that dropped in rankings and the "supplemantal" page that i believe could have somehow played a role in this.

As a result, the "main" page's impressions and clicks are now slowly returning to their old numbers. Whether or not there is any direct causation is, of course, still up for debate, but this is a very interesting correlation ¯\_(ツ)_/¯

[chaumi]

In case this might be helpful to some of you:

So I decided to consolidate the two pages - the main page that dropped in rankings and the "supplemantal" page that i believe could have somehow played a role in this.

As a result, the "main" page's impressions and clicks are now slowly returning to their old numbers. Whether or not there is any direct causation is, of course, still up for debate, but this is a very interesting correlation ¯\_(ツ)_/¯

It may be as straightforward as the content now added to the main page increased its relevance (in Google's eyes). ie it clarified or strengthened/supplemented the intent of the page. Or maybe that extra content has increased the relevance for subtle changes in intent.

I'd probably be looking at the differences between terms with impressions/clicks in its past life with what it's recording now. If they're the same terms then you probably did supplement the original content in such a way that it enhanced relevance for the primary topic. If they're looking different then possibly the way that G sees the page and its intent has been modified.

ofc, changes in other factors can cloud the judgment either way as well.

[edgarf76]

This is a tough one. I am going through this myself and am finding that consolidating to the review page is not always the best option. The supporting page may still perform better. The questions I ask is why the other pages are out performing and how can I consolidate. Good luck and please share your results. Personally, I like supporting pages. The internal linking can be very powerful and help rank both pages.