My top 3 pages make up 27% of my affiliate income

[bernard]

I just checked my affiliate income in the last 3 months.

Top 3 pages: 27% of affiliate income
Top 5 pages: 36% of affiliate income
Top 10 pages: 50% of affiliate income

I had aprox 100 pages bringing me affiliate income all together.

That's an eye opener isn't it?

There could be two ways to look at this. One, which is the classic pareto argument, is that I should spend 80% of the time on the 20% of the actions that create the greatest income.

Another argument would be that it's good that I have 50% of my income tied up in 90% of pages, since I am less vulnerable to specific changes.

What is your distribution?

 

[zentrader99]

I also have similar stats for most of my affiliate sites. I think 80-20 is more common for affiliate sites.

But, it does not mean if you had built these 10 pages website, you would have got similar traffic. I think Google picks some X% of the pages from each website passing a certain filter.

 

[bernard]

But, it does not mean if you had built these 10 pages website, you would have got similar traffic. I think Google picks some X% of the pages from each website passing a certain filter.

Definitely, content clustering is a major ranking factor imo.

I also see the many diversified income streams as a benefit. It's a big risk to be dependent on a few products or categories.

 

[Ryuzaki]

I did this exact type of analysis here: Overview of Traffic Expectations for Aged Authority Site (Graphs)

It features graphs and other data. It's not based on revenue but on traffic, but on this site I'm largely doing display ads so it's pretty correlated, around 75%, because I do get some affiliate commissions too, probably around 25% of the haul.

At this point, I'm focusing on making the problem worse for myself. I want more of these crazy posts that get huge traffic. Yeah, the others kind of raise the tide so the lowest dip of the traffic rollercoaster isn't so low, but they don't make the huge impact that these 80/20 winners do, and I'm not getting any younger.

I'll get back to raising the tide, but for now I'm aiming at the stars. Because at the end of the day if we're depending on traffic from someone else's algorithm, diversifying within their algorithm is only safeguarding you so much anyways. And I'd argue that having tons of traffic gets you more favor and gets you more signals to keep you safe anyways.

 

[bernard]

At this point, I'm focusing on making the problem worse for myself. I want more of these crazy posts that get huge traffic. Yeah, the others kind of raise the tide so the lowest dip of the traffic rollercoaster isn't so low, but they don't make the huge impact that these 80/20 winners do, and I'm not getting any younger.

I'll get back to raising the tide, but for now I'm aiming at the stars. Because at the end of the day if we're depending on traffic from someone else's algorithm, diversifying within their algorithm is only safeguarding you so much anyways. And I'd argue that having tons of traffic gets you more favor and gets you more signals to keep you safe anyways.

I know what you mean and I sort of agree.

I just feel like the chase for nr. 1 makes people generally do bad decisions, like buy links and spend all their budget on that. I've decided since early on, that I would win on content, quality and quantity.

The real interesting question to me is: How exactly do I replicate these posts?

I know you claim to have figured it out in your lab thread.

What I can tell is something along these lines:

1. Not extremely saturated topic
2. High quality and effort put into them
3. Revised several times with regards to content, UX, design, speed etc
4. Content clusters around the commercial content
(5). Targeted links

That's not very specific of course, but I have my own specifics for each of the above. It's not a coincidence, but it IS very time consuming.

I do think for my next Kitchen-niche site, that I will focus very strictly and strongly on those posts with expensive products and really deliver. I will choose only 10 product categories, 5 to begin with, and do everything above, with real product tests and everything. So I move that to beginning of a site process. Which is a hardcore process, probably takes 1-1.5 months with most of my focus going to that one site.

 

[Ryuzaki]

I know you claim to have figured it out in your lab thread.

I'm not trying to say I'm some mastermind. I just figured out something that's working on this project specifically that Google likes:

The update corrected some of the overreach, but I'm still way up. And I suspect that'll be how it works for each iteration. I've pumped out I think 3 or 4 more of these and have as many to go and I'm going to try to get links to them all and push this to the moon.

These topics are definitely saturated, but Google is liking my style over bigger sites attempts. I think the parts that are mattering most for me are:

1. Unique Layout
2. Higher CTR in SERPs

Everyone has the same basic layout, so I made mine different. It's nothing ground breaking at all. It's very simple and you wouldn't even spot it visually if you looked. The key is that it's different, and I think Google is measuring that as better, when other factors are aligned too, like..

Here's an example. If everyone has like "Top 100 Best Kitchen Appliances," I'm publishing "Top 10 Best Kitchen Appliances" because who in the hell wants to look at 100 items instead of just 10. As a reader I'd think the thought that goes into choosing 10 is more purposeful and critical than randomly slapping together 100. A bigger number is worse in this case. This is just an example.

The point is I'm trying to make it different and make it appease the users better, all the way to the title tag that appears in the SERPs. Then of course you have to get some exposure to them to help them pop so there's a compounding effect to all this, and in these cases that's meaning links.

 

[bernard]

Yes, I also am convinced I rank on user metrics more than links.

My fitness site exploded in rankings after Corona hit. I don't think it's because of traditional ranking factors, but because all the extra searches gave Google enough behavorial data, showing my sites superior, to push me up in the search. I forgot what that AI algo was called, that reshuffled the front page based on such factors.

 

[Ryuzaki]

Yes, I also am convinced I rank on user metrics more than links.

I wasn't doing any kind of misdirection when I said 1. Unique Layout as well as 2. Higher CTR. I think, in my case, both are playing equal roles.

John Mueller just went on the record in a Google Webmaster Hangout session and confirmed part of my thinking, where he said Google looks for "different" types of pages to rank. I'm trying to find a link to find the exact wording.

If I find it I'll share, but the point is that while everyone is trying to "match and exceed" Google isn't looking for more of the same. They want different stuff to measure and not fall into a local maxima when there's a much better type of result out there on the global curve outside of that local curve.

 

[bernard]

I see, ok, yes, I do that as well, with some of my "cheap product" pages, where I demolish the competition with barely any effort, except just having a radically different page. I should never be able to rank first for "cheap product" categories, but I actually do for a bunch of them.

I'm also good on "best product" stuff.

The only thing I'm struggling with is the broad single keyword stuff, which Google seems to prefer short articles in my SERPs, which has puzzled me.

They put in some of the long form pillar pages, but only a few, those with the most trust/links.

I'm going to try to figure that with the next site. I'm going to try to go more in the direction of Wikipedia, with a brief overview of topic, with links to all the subtopics. Not a category page, but a dense content packed page, with no monetization directly (maybe ads).

 

[eliquid]

John Mueller just went on the record in a Google Webmaster Hangout session and confirmed part of my thinking, where he said Google looks for "different" types of pages to rank. I'm trying to find a link to find the exact wording.

I'm not going to give anything away I know, but I have been practicing this for years and it works.

I was glad to see Mueller say that and confirm it even though I was getting success with it for a long time before.

My different type prob goes way beyond what others are doing, but I can confirm being different is key.

.

 

[bernard]

Regarding this being different thing.

I created a price scraper / calculator test site in Python and I've had several sales already ($50 each commission) with a month old site. I get highly targeted and valuable searches and my content is all auto-generated except for price data, which I do show with graphs and schema and all that.

This site would not have made a penny from Google if it had been a regular content site.