Irishwonder’s Black Hat SEO Blog

Recently, while looking at various stuff, I came across this:

While paid links is nothing new for the gambling industry and it’s not a secret to anyone that a lot of legitimate generally non-spammy branded gambling sites live and rank because of paid links of all kinds, and there are many legends out there as to the link buying budgets allocated to SEO campaigns in gambling, this is one thing that screams penalty like nothing else. Mass penalty even. Why?

So there is a service that decides they would cater to gambling niche specifically and launches an offer like this. Let’s do some maths. Let’s say they only end up selling 100-link plans so each link they sell brings them $20/month. That is $120/year. Now let’s imagine, just to simplify our calculations, that they only use brand new $10 .com domains for their network. Let’s imagine even that they do not use private whois. Further, let’s say they host their domains across several reseller hosts, $200/year each, with, say, 20 domains per host. That’s $10/year per domain for hosting. Let’s further imagine that they buy unique C-class IPs for each domain (that’s the point right?) at $1/year. So far, looks good, doesn’t it? They get $120/year off each domain and their expenses are $21/domain year. Wow you say, 500% ROI, what a business model, and rush off like mad to set your own niche link selling network. Of course this is a no brainer, you will be able to milk all those fat gambling budgets while providing an excellent and secure service, right?

Not so fast, I say. What have we forgotten? First: the content! You cannot sell links off empty sites can you? Let’s imagine you got really lucky and ran into a third world writer with decent writing skills whose jibberish articles don’t make a human eye bleed and this writer charges you a $5/article fee. How many pages of content do you need per site? Let’s say 10, you are greedy and you won’t update it too often. That’s $50/domain right on the spot. Our expenses now total $71/year per domain. Ok but we are still profitable, right?

What else are we forgetting? Links! To be able to sell links off a site, you need a site with at least half decent link profile itself. Even if you spam those links with machines, that’s still an expense. Say you hire some freelancer who builds you 1,000 links to each your site at $50/job (sounds like outright spam but ok, for the sake of simplifying our calculation let’s just live with this). Oh, but now our expenses total $121/year per domain! The setup is not profitable any more!

What I am trying to show here is there is no way a service like this will only be selling 1 link per domain per client site. Even cutting costs like crazy, they won’t make any profit (and I’m not even starting to pick at all the obvious flows in the above described setup). Hence, a service like this is bound to sell links to several client sites off each domain it owns and operates. And remember, they will all be in the same niche.

Now, why is this an important point to keep in mind? Here is why:

Now this is shocking, right? These sites have so much in common. And I didn’t even do anything special here, I just took 5 random brands ranking on the first page for “online casino”. How long do you think it would take Google to discover the whole network selling niche links? Even if they limit the number of client sites they sell links to from each individual domain, they are still outing themselves big time. To be at least somehow safe, a service like this would have to operate a really large network so that they could at least vary the sites between their clients enough to not be very obvious.