Off the Top of My Head

Manipulating Alexa Isn’t What It Used to Be

There was a time a couple of short years ago that asking what the importance of your Alexa ranking in a forum would get a million responses that it was worthless and all of those people were right. I then ignored those discussions over the years because they were all exactly the same. That was until a recent thread on Digital Point Forums where another new webmaster posed the question and although it elicited many of the same “It is easily manipulated” responses, there were a few dissenting voices. Now being used to post count padding, I wouldn’t normally have paid attention but those denying the ability to manipulate it are people who I regard as pretty up to date and knowledgeable in these areas.

So being drawn into the thread I decided to post my same response from a few years ago outlining all the reasons it could bot be trusted for rankings under 100k. I figured maybe these usually informed posters just didn’t realize the wide range of tools out there for changing your ranking drastically. The response back was a simple and plain one: If those tools work so well even now, prove it and manipulate your own rank.

Now it is really common for discussions in forums to break down when people just keep saying prove it but in this case I felt it was a legitimate point. It was so easy when I previously changed it, it would only take a week or two to do the necessary tests with minimal effort and be crowned king of that particular thread. I love proving peoples wild accusations based on myth wrong so this was a perfect opportunity.

So the experiment begins. The first step to any manipulation is to just install the toolbar on a few computers, preferably with different IP address’s. I only have two that fall into that category so I put the bar on my home computer and my work computer. I don’t normally put what is basically “spyware” on my working computer but I thought it would be beneficial to the challenge and the small tweak of the firewall would be worth it. Normally doing this and setting the soon to be manipulated site as the home page causes enough refreshing to give you a huge boost. After all, each install in the old days represented thousands if not tens of thousands of computers.

Then I pulled out the big guns. I rooted around on my hard drive and found an old Alexa booster program and set the machine going. I’m worthless when it comes to programming but as far as I can tell, it uses a proxy system to access your site thousands of times through the Alexa toolbar, “spoofing” a unique visitor each time and racking up the brownie points. Just in case that particular program was blacklisted by Alexa, I also found a few more boosters via eBay and installed those. I couldn’t wait to manipulate the ranking down to like 35K and show how ridiculous Alexa is as a ranking system.

Now normally I would see results at the next update which used to happen every Sunday night/Monday morning. I’m not sure when it actually happened but before I knew it, this site which had a starting point of 402k rose to 403k at the next update. To be honest I was expecting for it to drop by about 250k in one go like the old days so I figured the update didn’t take my work into account YET. So I patiently waited for another update and it went up again to around 404k. Now this was getting ridiculous as it seemed all my efforts had absolutely no effect. It wasn’t until a few weeks later that it actually settled back down to around 350K. You would be tempted to say that a 50K+ drop was phenomenal but when you are in the 400k range it is pretty much meaningless. To this point I still haven’t seen a drop down past the 100k mark which would have normally happened within a single update.

The testing I did convinced me that Alexa is not as easily manipulated as it used to be. In fact, it doesn’t seem to be able to be manipulated at all. One area where there does seem to be inconsistencies still is the fact that webmasters tend to have the toolbar installed in way bigger numbers then the average Joe. This is turn means that webmaster resources like directories tend to have highly inflated Alexa values. In the above example, the 400K site is actually a very successful site of mine that pulls in huge visitors and 6 figures a year. In contrast, if you are going by Alexa numbers alone, you would think that this blog has significantly more traffic and potential for revenue. This blog, however, is nowhere near the numbers nor success of the other site. What this means is although Alexa can be off when it comes to the reach, it is still a good comparison tool for sites in the same industry.

So if the above is actually true and I am of course now one of the believers, is the previous myth that Google uses a bit of Alexa ranking in its algo still so dismissible? If it is no longer so easy to manipulate and the only existing problem is comparing sites in different industries, it’s not a stretch that Google would use it, even if as a very minor part of the algo.

If anyone has a manipulation tool they believe still works, please email or post a link in the comments and I will definitely give it a try. I will only accept the methods though as anyone just posting, “It is easy to manipulate” will be ignored since I truly believe Alexa has either achieved a critical point when it comes to number of installs or have limited the number of same range IPs it will accept as votes.

This entry was posted on Wednesday, October 25th, 2006 at 8:22 am and is filed under Forums, Online Business, Search Engines. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.