Some Sandbox Clarifications

Apparently either more people read my blog than I thought, or there is a LOT of interest in the sandbox - (I suspect the latter).

There have been several people attempting to figure out how I was able to beat the sandbox, and since it's not really fair to announce an experiment and then say that the result is a secret, I'll open up a little here.

First, anyone who knows me knows I'm very free with information as a matter of course, but in this case I have been convinced by several people I trust that there is a significant commercial advantage to shutting my mouth (sorry - a guys gotta eat!). But I'll clarify some misconceptions and provide some hints, OK?

First, I'm currently redesigning my website for a reason completely unrelated to this experiment. I'm going to China this spring, and my Canadian Government clients are starting to want things in French, so I'm making the site international in flavor and adding several languages. Part of this process has resulted in my re-purposing the .us domain for a US specific area of my site. Therefore, following it now won't get you very far, directly. There will be some very interesting things indirectly, though. More on that later.

Here are the facts:

1. I have not tested this technique with a .com yet, so if there is a special rule for non-ccTLD's I have not tested it yet. I don't think so (it's more likely a competition issue) but I'm not sure. It's being tested right now, using completely different sites and domains.

2. There were no 302 redirects involved at all. Nor was there any scripting.

3. The technique requires one aged domain and one new one, but only one website is involved.

4. There are no methods used that are grey, much less black. No cloaks, IP detection, hidden redirects, etc. (I don't do that sort of thing).

5. I'm aware of 2 other methods that can deal with the sandbox - one using 302's invented by Scottie Claiborne and one involving a black hat technique. I'm not sure if anyone is using that one, but I'm even less likely to announce how to do it, as you can imagine! This is not an impossible problem, you just have to think things through. I have one more potential method that I haven't even tried yet. That's 4 methods that I know of already. I'm sure they are not the only ones.

6. One of the key issues to remember is that this is not some sort of evil plot to screw over new webmasters. It's a side effect of a relevancy algo. Remembering this helps you focus on where to look (and where not to!)

7. There is no magic bullet - no matter what method you use, there is work and planning involved. Sorry.

8. Hint 1: I discovered this while figuring out how to make websites localized.

9. Hint 2: although the sandbox affects entire domains, Google lists individual web pages. This discrepancy in behavior can be used to your advantage.

10. Hint 3: I had the original idea while writing my Redirects for SEO's series of articles.

That's all I can tell you. You now have 2 things in your favor: 1) you know it's possible (I didn't) and 2) you have the information above, which is a far smaller list of variables than I had to work with.

I'm interested in seeing who can figure this out, based on the above. Good Luck!

Ian

No comments: