So just last week I was trying to organize some stuff & I happened to come across an old Amazon affiliate niche website of mine which I’d kind have forgotten about… Well, I hadn’t really “forgotten” about it as such but I’d basically just left it so sit almost empty & pretty much unusable for about ~5 years…
And because I was bored (or should I say looking for excuses to procrastinate from the real work I was supposed to be getting on with) I decided to take some time out to fix it up & revive it… However when I took a closer look into it to my surprise I discovered that the website wasn’t quite as broke as I thought it was, and it had actually been generating content in the background for all of those 5 years… It just wasn’t displaying it… That was the only problem.
So it’s safe to say I was pretty excited by this discovery, on one hand to see if it could make me money & on the other hand mainly for the fact that it would make an interesting little case study as I was kinda curious how a site would perform in terms of generating search engine traffic when going from next to no content to over 100,000+ pages… Would it turn me into a millionaire? Haha – keep reading to find out… ?
Firstly – The Site’s Background
So just in case you don’t know I am actually kinda self-taught in web development… I mean I’m no expert at it & I’ve never studied it properly, but from what I’ve read/seen online I’ve managed to pick a few bits up over the years…
Anyway way back in 2014 I decided to build this little website that would pull content from various different sources & produce it all in an easy to follow feed so that people who were interested in the content could easily access it.
Now it’s worth pointing out here that I’m not going to share the exact specifics of the site – not at this point anyway – but what I am going to do instead is try to give you as good an idea as possible as to roughly how it works.
So yeah, it was a site that pulled content from elsewhere & basically reproduced it in a feed… Each day it would create a new page & each feed was dated by day – so as an example “niche keyword + 1st of January 1970“.
The idea was that all of these pages would get listed in search engines & then people would come to the site & click the Amazon affiliate links then buy something & make me some money in return…
However if you know anything about SEO you’ll likely know that due to the low-quality/duplicate content it didn’t really work as hoped (but back then I didn’t know an insane deal about SEO… I was more into the development side of things as opposed to the marketing side back then).
The good news though is that I had a plan B. I had also built an “auto-tweeter” for the site, so each day the content was published to the website’s Twitter profile with various hashtags & the site got a good amount of traffic from there.
Almost immediately I was seeing an average of around 200 visitors per day – and sometimes over 500+ depending who had retweeted the tweets during the day. The site was also getting great engagement too.
Sadly though it still wasn’t making a great deal of money as the products sold via the site were pretty darn cheap… So even though I was getting a decent amount of sales per day (which wasn’t bad for an automated site) it wasn’t really adding up to much overall.
The “good days” came from when people clicked the Amazon links & then bought something entirely different (like a dildo – yes LOL that actually happened)… But back in 2014 whilst Amazon was popular it was nowhere near as popular as it is today so still those “good days” were few and far between.
So What Happened To It?
Well to be honest because the site wasn’t earning me much money I didn’t really pay it much attention & instead I started to work on other stuff whilst just leaving it to do its thing in the background.
Occasionally I’d get the odd pay-check notification from Amazon & around £50 or so would get deposited into my bank account but one day back in around 2017 I came to notice that these had really dried up.
Upon looking it I discovered that the site had stopped working… All the content that had been published was still there but no new content was getting added so the homepage was just bare.
I’ll admit here that in the back of my mind I knew it’d likely be an easy fix but as I had other stuff going on coupled with the fact that it wasn’t making me much money anyway I just decided to pass up on fixing it.
That was it, I assumed the site had just bit the dust…
Fast Forward To 2019
So as I mentioned at the very start of this post, last week I found myself trying to organize some of my stuff. I have so many different websites & projects going on that I just wanted to clear the clutter…
In doing so I came across this old site that I thought was “broken” and I decided to take a quick look into it to see what had actually gone wrong with it… Which is when I found over 100,000+ pieces of “unique” content stored in the database.
Whilst the site had broken in terms of displaying the content, it hadn’t actually broken in terms of retrieving & saving it… I never in a million years thought that would be the case but hey-ho, it was.
So with 100,000+ pieces of content in the database I felt it was more of a duty to get it back online, just to see what would actually happen in terms of getting traffic (primarily from search engines).
Surely with 100,000+ pages you could catch some decent amount of visitors right?
Well that was the hope… And to give it the best chance possible I totally rebuilt the site from the ground up.
In doing so I learned that back in 2014 I really did suck at SEO (and web development!!) LOL.
Let me paint you a picture…
Not only did the site redirect mobile visitors to a separate “/mobile/” URL rather than being responsive (without any duplicate content protection), but it also basically just threw people straight to Amazon without making any attempt whatsoever at keeping them on the page or getting them to click through to others.
In fact probably <1% of visitors clicked to another page because the internal linking structure was essentially non-existent… It was just affiliate link upon affiliate link (whoops). ?
So I really went to town on it when I redeveloped it – I made it properly responsive & rather than affiliate links I put a massive emphasis on internal linking – and on top of that I made sure every page contained a relevant video to maximize time on page.
The structure remains the same in terms of targeting keyword+date like “niche keyword + 1st of January 1970“, however rather than simply firing people off to Amazon directly from that page they are now taken to individual product pages on the site which contain more information pulled from various 3rd party sources.
So the structure on each product page looks like this:
- Brief Info
- Affiliate Link
- Comment Section
All of the sections are pulled from different sources and compiled together – something that is not done anywhere else, so whilst therefore the content may not be unique, the site still provides good value to the visitor.
In fact to be honest, I will say that the site provides massive value. It’s very useful, even though it is auto-generated.
But There Are Some Problems…
One of the biggest problems is that the site’s picked up some pretty dodgy backlinks over the years, which on it’s own wouldn’t be too much of a problem… But coupled with the fact that it’s been sat empty (and full of affiliate links on the pages that do exist) for so long it likely looks very spammy in the eyes of search bots & the domain reputation could be kinda tarnished.
Don’t get me wrong though it’s picked up some pretty good backlinks too – with some even from Wikipedia, so I think that speaks volumes about the quality of the content of the site, even if it is auto-generated.
Now I know with persistence the domain reputation can be remedied but the reason I’m mentioning it here is because it might see things getting off to a slow start – and the domain is an exact match for the keyword (which has 100k+ phrase match searches per month) so I really don’t want to change it.
The other problem – well, what I see as a problem – is the sheer amount of content getting added.
Going from next to no content to over 100,000+ pages of content will likely be enough to make the search engine robot’s eyes pop out when it discovers it… And whilst I could have staggered it I thought what the heck, let’s have some fun.
After all where would the excitement be in a case-study titled “I added 10 pages each week to my niche site, see what happens“…
“I just added 100,000 pages” sounds way more fun right? Heh ?
IMO The Positives Outweigh The Problems
Yeah the problems kinda suck & having a domain with a tarnished reputation ain’t no fun when you’re trying to rank a new site (though hopefully the fact the site never went down altogether will help it out)… But there are some pretty freakin’ good positives.
First off, keyword I’m targeting gets 100,000 phrase match searches every single month.
Drop 1 of the words out of it and it gets 1 million phrase match searches a month.
Okay, sounds good – but the competition must be real strong right? Surprisingly, nope.
In fact in total there are only 102 different websites competing for the keyword…. Pretty crazy huh? However as always, the site taking the top spots is a pretty high-authority site so it’d take some beating… But it can be done!
And hey, with 100,000 matches I’m not gonna lie I’d be pretty happy at sitting position 2 as well. ?
Oh yeah and the other HUGE positive is that the site is totally freaking automated… So if this works & it gets a decent amount of traffic then the income generated will be 100% passive. Plus now that Amazon’s more mainstream I’m hoping that the potential earnings will have increased anyway – so even if I can get it back to the traffic that it was previously at I’ll still be pretty happy.
But come on – 100,000 pages of content… If it pulls off & they actually get indexed (which they should) then it’s safe to say that I’ll probably be sitting hoping for a few more than 200+ visitors per day. I mean even if only 1% of the content gets clicked per day that’s still 1,000 visitors which is quite a whopping number – considering they’ll all be super-targeted.
Yeah I’m Excited
Even if the site doesn’t go anywhere at all, no big deal – it took me about 2/3 days to fix & setup… But all the same I love experiments like this & it’s the first time I’ve ever truly thrown such a large amount of quality content at a search engine so I’m really intrigued to see what happens.
And yes, I said quality! I know it’s auto-generated, I know it’s sourced from elsewhere (not unique), but honestly the value it provides to the visitor as far as I’m concerned is currently better than any of the sites out there battling in the same niche.
But will the bots be able to figure that out? That’s the important bit. ?
Now For The Case Study Part…
You wanna see the numbers right? Okay, so here – first of all this is what the site was looking back in it’s heyday in 2014:
You can see from its launch in June of 2014 that the website quickly grew (primarily through the use of auto-tweets), peaking at 678 users per day shortly after in the following month… Which is pretty impressive!
The site managed to sustain around 300 users per day after that, but in 2017 the Twitter account was banned (apparently Twitter changed their automation rules and I missed their mail) so the traffic went on a general decline… But then also that year as mentioned earlier in this post the site also “broke”, and so the traffic plummeted to next to none as shown below:
And now from March to April of 2019 (right to the day of writing this post) this is how the traffic is looking:
Peaking at around 9 users per day the traffic is pretty much non-existent, and the spike (then sharp decline) towards the end is just when I started working on it… And then finished. Most of the “users” are actually just Facebook bots as I was using the Facebook comments plugin & for some reason that seems to bring a lot of Facebook bots with a “?fbclid” query string in Google Analytics.
I’d view the page, Facebook would detect I’ve loaded their iframe & then it seems they’d send a bunch of bots to the site from all different locations around the world. No idea why, but yeah…
Now Watch This Space…
The site’s finished, all the content that I found in the database has been added & each day it’s doing its thing in publishing more & more… So from here it’s just a waiting game to see what happens.
It was Friday the 5th of April when I finished the main changes & there’s been no reflection in the search results as of yet (4 days later) – but of course no doubt with such a large change the bot is probably taking some time out to think “what the heck?“.
Maybe the site will just be ignored – maybe it will be de-listed on the assumption of spam – or maybe they will all rank and make me an overnight millionaire. ?
The latter would sure be nice but only time will tell, so we will wait & I will update this post as soon as there’s anything worthy of reporting. ?