How to get 9000 free links on autopilot

closeThis post was first published 2 years 5 months 9 days ago. It might make references to techniques or tactics which no longer work or which I no longer endorse so please proceed with caution.

How’s that for a spamtastic title! This is going to be a fun post about how I accidental built 9000 links from over 400 domains with virtually no work. You probably won’t want to do it for yourself but you might just learn something.

About a year ago in the early stages of setting up my first blog network I built a little article directory site. The idea was to use it as a sort of private ezine articles. Anyway I must have got sidetracked or forgotten about the domain because I never touched it again until about a month ago when I started getting emails from cpanel telling me I was running out of bandwidth on the server this thing was setup on. Strange I thought given its a site with no content and I presumed no traffic. So I checked out the site and to my surprise I did have content – 150,000 pages of it in fact…


As you’ve probably guessed I hadn’t made the registration private on the article directory script and the bots had been busy submitting thousands of articles to my directory on autopilot on all sorts of fascinating subjects like…

“Why it’s possible to Discover out Males Ebel 1911 Discovery Chronograph Rubber”

Now that’s what I call a headline! Fair play to them. What interested me more than the content though was the sort of links the site had “attracted”. According to SEOmoz 8988 links from 436 domains at the last count.

Now as much as I’d like to think the quality bot spun articles about herbal penis enlargement treatments had attracted those links naturally, actually what’s going on here is the bots are shooting a bunch of links back to the articles they’ve posted on my site from other article sites, xrumer powered forum profiles, blog comments, hacked sites and so on to get the articles on my site indexed. Think of it like a “link wheel” on bot steroids. While most SEO folk only started talking about link wheels in the last 20 minutes the black hats have been using the same methodology for years and as with most of these things I reckon if you’re able to automate and scale it enough its probably still going to work to an extent.

So what’s the lesson here? Well there’s a few. Firstly if you want a shit load of free content and links then all you’ve got to do is setup an article script and get yourself on an article demon directory list and the links will come to you! They won’t be great links but some links are usually better than no links (not always), so you might be able to do something with that.

But more importantly what this little anecdote tells me is you really have to take with a liberal pinch of salt what you see at a glance in your link research tools. I mean, how many link builders out there would take a glance at a link profile like this and think.

Yeah, ok, that’s not too bad a site. I’ll shoot them a link request/ pay for a link/ submit a guest post over there.

We’ve got lots of links from sites with decent domain MozRank….

Not great page authority but not the worst I’ve seen…

A nice mix of themes…

Even a scattering of good old fashioned PageRank. Not a lot, but better than many legitimate blogs I look at.

Actually the only at a glance metric which flags that there’s something not quite right about this link profile is the Majestic ACRank score…

I don’t usually pay much attention to this metric but in this case its linear counting system actually gives the best insight into the true state of this truly horrible backlink profile.

So here’s a few takeaways:

  • Content and links are not as hard to come by as you may think
  • But not all content and links are created equal
  • Link research tools can’t detect spam
  • So when doing link analysis look at as many data points as possible
  • ACRank can be useful
  • Before you even think about approaching a site for a link, actually check out their backlinks by visiting the pages which link to them
  • Bear in mind that most link brokers will just look at the raw metrics and won’t do these checks so if you’re buying links or outsourcing link building make sure you’re doing due diligence
  • If you just trust what you see in your toolbar you could end up getting a load of links from sites like my little article directory experiment, which probably won’t help you rank that much!

7 Comments How to get 9000 free links on autopilot

    1. Geoff Jackson

      @Mania: The links would a) not have been built all in one go and b) would not all be pointing to the same page on John’s article website. The likelihood is, that most of the links were actually manually processed, with each ‘link builder’ auto publishing to a range of articles they submitted across a range of article directories to create the link wheel effect that John discusses.

      Backlinks from article submission websites have already been devalued throughout the early stages of Google’s Panda updates. Any large quantity of backlinks being processed and indexed to the same page at the same time will raise flags at Google, and could subsequently be discarded or the domain penalised but if any of the backlinks circulating aroud the link wheel can be easily associated to bad neighbourhoods, then it won’t take long for Google to slap the majority of the domains participating in one go.

      Reply
    2. Ando Hiroshige

      Yes and no. Those links are used for “link wheels” and not direct linking. The premise of it it is that even if google penalizes those pages containing spun content, link-juice is still passed on to the pointing domain. Of course this is debatable :/

      Reply
  1. Erica

    Woohooo! That’s so weird! How is this site doing after the Panda update? How much traffic do you get from it? Is there a lot of duplicate content? We want to hear more about this case study!

    Reply
  2. Scott

    I also wonder if Panda will be tough on such sites going foward. On the other hand, it sounds like a lot of content for very little effort.

    The real test will be how much revenue and audience it produces.

    Reply
  3. Roberto

    Is the site still ranking well? Just wondering if so many links actually helped increase it’s pagerank and if it did, did that have any effect on Google search results?

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>