Another Reason To Not Use Alexa Statistics: They’re Too Easy To Manipulate
By Collin Maessen on comment
I’ve already written several blog posts about Alexa and how notoriously bad its statistics are. Not really surprising that they are bad if you know how the data for those statistics is gathered. For the bulk of their data Alexa relies on people installing the Alexa Toolbar (or a toolbar that passes information to Alexa). It’s this toolbar that monitors to which websites you go and how you end up on them.
This has as a result that demographic, used browsers, and even the country visitors are from influence the data that Alexa gathers about a website. This can introduce serious artefacts and biases into the collected data and basically makes the statistics generated from it worthless. At best it can give you an idea about how well a website is doing, but that doesn’t mean that what you’re seeing matches reality. I’ve already written a far more detailed blog post about how Alexa works and why you never should rely on the data it provides; it’s just too unreliable.
The best illustration I have about how unreliable Alexa’s statistics are is from when I compared HotWhopper and Real Sceptic. According to Alexa I had a higher ranking than HotWhopper and HotWhopper’s Alexa ranking was dropping noticeably. But HotWhopper had fourteen times the number of pageviews and had nine times more visitors than Real Sceptic. What made it even worse was that HotWhopper’s pageviews and visitors were increasing. Statistics from direct measurements told a very different story than the one Alexa was telling.
It also made me wonder if it was possible to influence Alexa ranking/statistics. As Alexa does mention on their website that there are some other sources for the data they gather. Turns out this is surprisingly easy to do.
Feeding data to Alexa
One of the first tricks I heard about that could influence your Alexa ranking was by using the Alexa widget on your website. The reasoning behind it was that Alexa then can use it to collect more detailed and reliable data. Sounds logical and reasonable, so I tried it.
During a period of over half a year I monitored how either adding or removing the Alexa widget influenced my Alexa ranking:
- 2013-08-14 – Added hidden Alexa widget
- 2013-12-02 – Removed hidden Alexa widget
- 2014-02-22 – Added the hidden Alexa widget again
- 2014-03-16 – Removed the hidden Alexa widget again
Here’s my Alexa ranking during that period with the last 3 widget changes marked:
I couldn’t mark the first change I did as the graph doesn’t go back far enough. But in older graphs that I do have you can see it triggered a steady rise in my ranking. Though this rise was really close to an increase in traffic to my website, so I wasn’t sure if the Alexa widget was influencing my ranking or how much. That changed when I started removing and adding the widget.
The first time I removed it (first arrow) it didn’t seem to have much influence on my ranking, it continued on for a bit before it suddenly stopped. Between that point and the second arrow my traffic tripled, yet Alexa didn’t show any hints of that. Just a couple of small blips and dips that vaguely resembled some traffic changes. But no sign that my readership was growing.
When I turned it on again my ranking suddenly started plummeting. Really odd as that was during my best month so far with the amount of traffic my website was receiving. It gets even stranger when I removed the widget again in March. During that month I didn’t publish new content, due to that I got the same amount of traffic that month as in December of 2013. Which is a third of the traffic I got during February of 2013. Yet my ranking went up.
How the Alexa widget will influence your ranking seems to depend on what Alexa thinks is happening on your website. Thanks to that your mileage will vary, but it does look like your Alexa ranking can be significantly influenced by it.
Search engine gaming
This one I stumbled on by accident, but I noticed it because it did influence the Alexa statistics for Real Sceptic.
One of the ways you apparently can influence the data Alexa gathers about your website is to write content about Alexa. What then can happen is that people who are looking for information about how Alexa works end up on your website. It’s no surprise that people who look for that kind of information have the Alexa Toolbar installed. As they belong to a group of people who are most likely to be interested in learning more about Alexa and how it works. The effects this had on the statistics for my website is rather obvious when you see the search keywords Alexa reports:
Sure, I do get enough traffic for my Alexa content. But there’s also other content on this website that gets a decent chunk of the traffic from search engines. But the keywords visitors use to find that content isn’t present at all in the Alexa statistics. Yet I get more traffic for that content than for my Alexa content.
But it doesn’t have a big influence on my Alexa ranking. It’s certainly there, I can see changes in ranking that matches the amount of visitors I got on my Alexa content. But it only works if you rank high enough in search results. And then it’s only the cherry on top with the other methods available for gaming your Alexa ranking. So it’s not the most effective one, but it does help somewhat.
Alexa Toolbar shenanigans
This is a big one. If you are unscrupulous enough and manage to pull it off, it will have a big influence on your ranking.
One of the best ways to increase your Alexa ranking is to get your visitors to install the Alexa Toolbar. If they visit your website often enough and/or you have a big enough audience this will increase your Alexa ranking significantly. I’ve seen this happen several times now and it works.
However, you are asking your visitors to install spyware on their computers. Not something I condone or recommend doing. But despair not, there’s a second method that’s almost just as good as getting your visitors to install the Alexa Toolbar.
Using the Alexa Toolbar yourself
Yes, installing the Alexa Toolbar in a browser that you use for your website is ludicrously effective. How effective you might ask? Well, this effective:
Sou who manages the website HotWhopper installed the Alexa Toolbar in a browser she only uses for HotWhopper. With that she managed to change a downward trend, despite getting more and more readers, into an upward one. It’s so effective that she now has a far better ranking than she had before her ranking started to drop.
So if you’re willing to put up with using a different browser that uses some spyware this is the way to go to help your ranking.
Conclusion
I already considered the statistics generated by Alexa basically worthless, with how they’re gathered they never could be accurate. I only considered them somewhat useful for getting an idea about the traffic a website gets. Although not a reliable method and not suited for comparing websites. But now that I’ve seen how easy it is to manipulate the Alexa statistics I consider them worthless. You just don’t know if someone is actively manipulating the data Alexa gathers for a website.
The advice I can give you is to not rely on the statistics Alexa generates. But maybe you can have some fun with it trying to manipulate your ranking. I certainly had during my experiments.
Updated 2014-06-03 @ 19:50
Decided to run a test with the Alexa Toolbar in a browser that I used solely for this website, here’s the result:
My website rose 150,000 positions in just a couple of days. Not as dramatic as for HotWhopper, but a nice result for very little effort.
Featured comment
-
Interesting bit of detective work, Collin. I still go to HotWhopper with the Alexa toolbar browser, not as often as I used to but still daily, usually.
In my case, the best indicator of what’s happening re Alexa is to see how HW ranks in country ratings. Back in March, HW got close to ranking 4000 in Australia – and Alexa showed it as more than 80% of visitors from Australia, which isn’t so. I figured that was mostly my couple of toolbar clicks each day. Makes you realise what poor info it relies upon for its rankings – and probably how few people in Australia use the Alexa toolbar. (I know where the visitors come from – 40% from N America, 33% Europe, 23% from Australia and NZ, and the rest from Asia and Africa and elsewhere.)
It’s still highly ranked in Australia but has dropped back to around 8000 now and is sitting between 50% and 60% from Australia, so I figure there must be visitors from elsewhere who have the toolbar installed. (Actual monthly visits have gone up about 30% since December last year, when that chart starts – according to Google analytics. But I doubt Alexa would know that:))
Collin Maessen is the founder and editor of Real Skeptic and a proponent of scientific skepticism. For his content he uses the most up to date and best research as possible. Where necessary consulting or collaborating with scientists.
Interesting bit of detective work, Collin. I still go to HotWhopper with the Alexa toolbar browser, not as often as I used to but still daily, usually.
In my case, the best indicator of what’s happening re Alexa is to see how HW ranks in country ratings. Back in March, HW got close to ranking 4000 in Australia – and Alexa showed it as more than 80% of visitors from Australia, which isn’t so. I figured that was mostly my couple of toolbar clicks each day. Makes you realise what poor info it relies upon for its rankings – and probably how few people in Australia use the Alexa toolbar. (I know where the visitors come from – 40% from N America, 33% Europe, 23% from Australia and NZ, and the rest from Asia and Africa and elsewhere.)
It’s still highly ranked in Australia but has dropped back to around 8000 now and is sitting between 50% and 60% from Australia, so I figure there must be visitors from elsewhere who have the toolbar installed. (Actual monthly visits have gone up about 30% since December last year, when that chart starts – according to Google analytics. But I doubt Alexa would know that:))
It wasn’t that much work to gather all this information, it just takes a while to find trends/effects. I already mentioned that I was doing this, but when I did I didn’t have all the information I wanted yet.
I’m tempted to use a browser with the Alexa Toolbar installed just to see if I can catch up to you. For some odd reason I’m never that far behind your site, despite you getting more traffic. I would just need to raise my ranking by 300,000 to do that, which should be doable with that method. Would be hilarious if I managed to get into the 100,000 top websites.
Very nice article on Alexa Ranking and statistics, and whether it really matters at all.
For me personally, I have installed the Alexa Website widget and I am using the Alexa Toolbar, and no matter how active I am on site and off site, my Alexa ranking and site links are frozen…there has been zero change…literally
Maybe I’ll do as is mentioned in this article, and take the widget off, toss away the toolbar for a while, then wait and do a test a few months later and see what data is shown….
.
Good article on alexa ranking and alexa toolbar.