Eli wrote about desert scraping, which is the practice of digging around expired sites on archive.org, grabbing their content and reposting it on your own sites. Cloning expired sites is a similar idea that I had recently while talking to my friend, Slightly Shady SEO. The difference between desert scraping and cloning is that with cloning, you register an expired domain and replace the content that used to be on it. If you enjoy this, pleaseÂ subscribe to my RSS feed – you’ll find many more ideas like this in my greyhat, link building and other archives.
(I called the idea cloning because you’ve got the exact same genes as the original – domain and content, maybe even hosting if you’re clever – but you’re not the original.)
How Desert Cloning / Cloning Expired Sites Works
You don’t do cloning with any old domains – you do this with expired domains that are still listed in the Yahoo directory (or DMOZ or insert name of the handful of directories worth submitting to). If you’ve found expired domains with links from other sources, those might be good candidates for cloning, too.
The reason you replace the content on the original domain is that you want the links – the more powerful the directory, like Yahoo, the better – to the domain to stick. That’s the whole point of buying these expired domains, to begin with – they’ve already got links pointing to them, maybe even some pages still in Google’s index if they just recently expired.
Imagine you’re an editor reviewing the sites in your directory. If they’re the same as what you saw last time or the same as what you have in the cache, there’s no foul. So you leave the link as is. Ownership change in the registry shouldn’t make a difference – the new owner could be a brother/sister/relative/business partner/same person with a legally changed name or newfound desire for privacy.
Once the Old Backlinks are Sticking…
Well, actually, you don’t clone the sites exactly. You throw in a link or two on the cloned site to your other properties, eventually. “Eventually” taken here to mean, after enough time has passed and you feel confident that, if there was any regular review by directory editors of sites whose domains had expired, that said review has been done and you’re not getting booted from the Yahoo directory.
(Recall that Rand tried and couldn’t because he didn’t want the meta description being pulled from it for SEOmoz. I can’t find the link to Rand’s post but if you have it, please drop it in the comments and I’ll update the post.)
So eventually, you slip in a link to your site(s) or client’s sites. If you’re a blackhat, I’d imagine you would point this to a mid-level site in your network; not to your foundation level spam but not to a money site either.
The Blackhat Appreciation of This Tactic
If you’re a blackhat, your goals for any particular site are short-term. You might have long-term goals like completely automating a particular form of spam, or some of your processes. But mostly you want your money site(s) to rank for a few weeks, pull in some cash and move on. (Complete automation – even for blackhats – is a pipe-dream IMHO, but that’s another post entirely; on a related note, I said [whitehat] SEO can’t be automated before Loren did, yet he got all the attention. Sigh…)
So as a blackhat, you don’t really care about defensible [here taken as non-SE dependent] traffic (if you, the reader, do care, find out if your traffic is defensible here). You aren’t building a brand or a legitimate business.
(Yes, I realize the average blackhat makes 10x what I do, by legitimate I mean what the average person considers respectable. Plagia “Scraping” isn’t usually what people consider respectable, even if it looks like one of those boards with plenty of copy-pasted words on it in the end, and not like it came off some particular person’s work.)
And that’s where things break down between blackhats and whitehats. That’s why whitehats might not be so well-advised to use this blackhat tactic.
The Whitehat Approach Focuses on Longterm Value
A link, or even a few of them if you clone many old sites in the Yahoo directory, might help you a bit. But the problem is that it’s not defensible. It doesn’t have long-term value independent of the SEs. It’s unlikely to send any traffic if we’re going to use the ultimate litmus test of a link’s value. It’s not defensible.
Suppose that the algo changes tomorrow and resets expired domains at 0 (Eli’s got some anecdotal evidence they reset, but not entirely back to 0)? Or in a more likely scenario, suppose that rather than such a drastic change, your sketchy practices get hit by a filter. Congratulations, you just wasted your time and [at least some] resources. No problem for a blackhat. Churn and burn is part of the bh game. Besides, you’ve had adsense there for the site to pay its own rent, as Eli would put it.
Crummy situation for a whitehat though. You can’t rely on those links’ value to remain constant, so you can’t rely on them longterm. Because engineers care about the means as much as about the process, and the means of getting these links here are debatable.
How Greyhats Can Use Cloning To Their Benefit
If you have a higher risk tolerance and don’t mind grey hat (not to be confused with Grehan) tactics, then you can use your clones to link to the money sites or mid-level sites until you get enough link juice from elsewhere. So you can salvage some value from the time you just spent reading these ideas on cloning.
What’s a mid-level site? I’m borrowing yet another idea from Eli (really, you should subscribe to his feed) and repurposing it. A mid-level site in Eli’s world is decent quality spam, if that makes sense. Not the blackest stuff, but not such clean snow that you’d eat it either. It serves as a buffer between your money sites and your crummy auto-generated spam, to avoid tainting the money site by association with a bad neighborhood (Remember Google-bowling? Some spammy links will accrue to any and all sites; mid-level sites prevent a money site crossing the threshold.) A greyhat can use the idea for hiding his backlinks amongst a distribution of sites they control.
Let me give an example by way of illustration. Suppose I operate a shoe store. It’s a competitive industry, and I’m not interested in the competition copying my links (which tactic I outlined in this link building post and this one, and which ideas this smart guy integrated with SEO Quake). What do I do? I create a set of buffer sites.
The check-out-the-competition’s-backlinks-and-copy-them tactic is typically done by looking at the first level of results for a linkdomain: search. That’s because if a site is ranking, it usually has thousands of links and it isn’t practical for a human to look at the links pointing to the backlinks. Short of setting programs to the task of crawling the backlinks of a site’s backlinks, and the backlinks’ backlinks’ backlinks, humans can’t keep copying your links when you control the first layer.
Obviously you won’t control the whole first layer – other people will be generous and link to you for free. But when you know you’re going to get a juicy link (by guest posting, for example, or with user-generated content that’s going to get some nice attention), you can direct that at one of your mid-level buffer sites. And guess what? No matter how much they beg the firstname.lastname@example.org for the same link, your competition won’t get the link ;).
How Stupid Greyhats Shoot Themselves In The Foot
One more note about buffer sites: Don’t be stupid and rank them. If they rank, they’ll have their backlinks checked out just like your own site. Hell, keep them out of the top 100 if you can. You just want them to have some Pagerank (not my understanding of it, based on what these two dolts thought, but rather Matt Cutts understanding) and pass that on like a nice link or two.
Conclusions On Cloning Expired Sites
The bottom line is that building backlinks from cloning is alright, but certainly not a longterm solution for links. The reason you shouldn’t rely on them, at least not exclusively, is because they’re not too hard to find (if a blackhat can program a way to find these domains, the combined PHDs of MSN, Google, Yahoo, etc. can likely write the program too). Whitehat sites shouldn’t be built on shaky foundations. If they get hit by a filter, that’s a tough blow to absorb when you don’t have thousands of auto-generated sites.
What this technique can be used for is to build links for your buffer sites. If you’re working on that level of link building already, then congratulations. Otherwise, you better hope that the competition stays weak and ignores competitive SEO linkbuilding strategies for the foreseeable future. But that’s not defensible ;).
If you liked this item on advanced SEO ideas, you’ll probably enjoy the rest of my blog too, so you should probably subscribe to my rss feed. this item on using Google Trends for domaining and expanding your keyword targets, and perhaps have an interest in my Facebook poll revealing that Facebook users don’t know that the Facebook blog exists. And likely the greyhat (grayhat?) SEO archivesÂ as well, which feature such ideas as disguising text link ads as AdSense.