The mystery of Alexa Rank: Users grow but rank decreases?

Most of you probably already know how Alexa rank is calculated. However, there are so many people who still have no clue about it, and they don’t know how really it is calculated. We also have an article about that posted some time ago, and we’ve explained everything in details. So everything should be clear about it, isn’t it? No, there’s still a small mystery about it and this is where we’re going to talk about it.

We did some testing like 5-6 months ago (with multiple websites), and the results are a little bit of mystery to us. Or it’s just another solid proof that sometimes Alexa rank can be very inaccurate and here’s why.

So we’ve tested 4 different websites aka sites. At first, each website had Alexa rank above 2 million. We’ve sent the same amount of visitors (with Alexa toolbar) by using our secret technique (won’t tell what technique, because it’s secret), and the visits were real and unique (visits were in 3 digits). Each website was updated once a day with one new article. We did this for 2 months and later increased (almost doubled) the amount of visitors only for 2 websites. Another 2 websites received the same amount of traffic. We did this for another 2 months.

After this test, the results were kind of silly and here’s why:

At first (in first 2 months) rank was constantly dropping for 3 websites but somehow was increasing/decreasing for 1 website. Remember, the traffic was the same for all and all were updated the same. When 3 websites already had Alexa rank of less than 400,000 (1 actually had less than 200,000), that another website, which had some ups and downs, had it above 700,000. Basically this makes not much of the sense, because each website’s rank should’ve been almost the same or at least way closer to each other.

After we increased traffic for 2 sites and other 2 sites received the same amount of traffic, we saw even stranger results. 1 of 2 sites with increased traffic went below 100,000 (without any ups and downs), but another was going up and down and got stuck between 100k and 150k. Another 2 sites with the same traffic had different ranks as well. 1 site stayed between 200k and 300k, another was bouncing between 400k and 500k.

What was the strangest part about this that 1 website (with increased traffic) was below 100,000 and we almost doubled its traffic again. Its rank started to decrease (was going up), it even went above 200,000 and was bouncing between 200k and 270k for almost 3 weeks. Later we increased its traffic again, and its rank started to increase, but even with increased traffic, its rank was bouncing between 120k and 160k for almost 5 weeks.

So what this proves that Alexa rank can be way too much inaccurate, and you really can’t take it seriously at all. And the worst part is that a lot of webmasters still take it seriously.

Now, you’re probably thinking that our technique had issues. No, all visits were from different computers with Alexa toolbars installed in browsers. All visits had the same duration of time spent on landing page. Alexa rank was decreasing successfully. So our technique was working great.

So the whole point of this article is not for showing how terrible Alexa rank is or anything like that. The whole point of this article is to show people that they shouldn’t take Alexa rank seriously. Again, a lot of people still take it seriously and they worry about it too much. Maybe this article can help to reduce the amount of people who still takes it seriously.