Amazon's Alexa Traffic Rank is Biased. Increases Ranks For Sites That Exploit Toolbar Users.
By
Staten Island, NY Posted: 12/31/2014 1:00:00 AM
When Alexa launched, I was excited that somebody created a traffic ranking system for the web. As it turns out, the whole system of increasing your Alexa Traffic Rank is so completely biased, it makes you wonder why it still exists.
When I say biased, I'm not just saying it's a little biased. It's a completely biased, and the funny thing is that everyone knows it, and that's why so many people LOVE Alexa. The bias is so obvious that there are now people making money helping sites exploit the bias to improve their Alexa Traffic Rank. In fact, Amazon is cashing in on it too.
What is the Alexa Traffic Rank? How is it biased?
Before Alexa first launched, there was no system to measure the popularity of a web site, so Alexa decided to borrow the concepts behind TV ratings systems pioneered by AC Nielsen, by polling small groups of people then scale the numbers up to get a general idea of what the masses are up to. Sounds simple enough, after all, that's how polls are done, and those are pretty accurate. Right?
If you were doing a political poll, and your sample group consisted 1000 people, would that be enough to determine the results of an election? Maybe... but it depends on who's in the sample group. If the sample group consisted of just the delegates at the Republican National Convention, would you still think the results are accurate?
That's where the bias comes into play. Amazon's Alexa Traffic Rank is based upon statistics that Alexa gather from two major sources. The first is results measured by Alexa's Toolbar and Browser plug-ins like the Google Chrome extension, the second is by getting site owners to include tracking code on their site.
The idea behind the Alexa Toolbar and browser extension is simple, you download the toolbar or install the plug-in, and Alexa tracks every site you visit, including how often you visit, what pages you look at, and for how long. They send that information back to Amazon and they gather the information into a big database.
Sounds wonderful. If they poll enough people, they will know which sites people visit most, and spend the most time on. It may not be perfect, but as long as they have enough people using the Toolbar or Plug-ins, what could be wrong with that? Everything.
It's even worse than polling just the delegates at a convention, because the it's loaded with sites exploiting the system to skew the numbers, by simply attracting (or creating) visitors who are using the Toolbar or the Plugin.
How sites take advantage of the bias.
Let's say you just launched a new site. On day one, you literally have zero visitors. As the site developer, odds rare, you are one of the relatively small percent of users who actually use the Alexa Toolbar or one of the Browser Plug-ins or Extensions. So, when you visit your own site, congratulations, you now have at least one confirmed visitor. Like a single vote in a poll or an AC Nielsen Family, your own visit now represents a much larger number of potential visitors
For some site owners, if you didn't have the Alexa Toolbar on your own browser, Alexa wouldn't even know you had even one visitor... ever.
Hopefully you are starting to see the bias. It gets even more interesting when you take it to the next level.
So, as a new site owner, who's smart enough to install the Alexa Toolbar or the Chrome Extension, Alexa now sees you, and knows you visit your own site, and you now know that since some sites are run by people without those tools, those sites remain essentially invisible to Alexa, with zero traffic.
Based upon that alone, what would you do next? Look for more Alexa Toolbar users?... or Even better...
The Alexa Traffic Rank Snowball Effect... When your existing audience starts using the toolbar.
Yes. It's that simple for sites with almost zero recorded traffic on Alexa to suddenly get hundreds of registered visits each day, making their site immediately rank higher than a potentially more popular site that does not have any users who use the Alexa Toolbar.
Another boost... creating content that attracts existing Alexa Traffic Toolbar visitors.
Now that you see that it's easy for a site to suddenly start getting registered visitors on Alexa, there is another clear way sites are exploiting the bias, by getting people who already have the Alexa Toolbar to visit more often. This sounds difficult, but it's not that complicated.
As I mentioned earlier, although the bias is well known, the truth of the matter is, site owners love the bias, specifically because they know how to exploit it. An odd side effect is that a large percent of the people who use the Alexa Toolbar are site developers, who check their Traffic Ranks on a regular basis. These Savvy users often search the web for for SEO tips, and therefore visit sites that of interest to site owners. Guess what? A lot of those site owners already have the Toolbar... and are now logging traffic to your site.
How Alexa tries to eliminate some of the bias... but fails.
A non-developer wouldn't even dream of installing the Alexa Toolbar, because it doesn't serve any purpose to anyone without a web site, so they let site owners customize it to include things their audience may want. The thought behind this is that will create a larger pool, that includes more non-developers.
The trouble is, it still skews the audience to favor sites that cater to or exploit Alexa Toolbar visitors, while lowering the Traffic Ranks of sites that are not savvy enough to either encourage or attract Toolbar users.
The Second Bias... sites that PAY to DIRECTLY track their ACTUAL visitors
As I mentioned earlier, the second way that Alexa tracks traffic is by getting sites to pay them a monthly fee to insert tracking code into their web sites so that they can directly track the ACTUAL daily traffic.
Well, that sounds terrific. It's no longer a guess. What could be wrong with that?
To start, that just changes the bias. Instead of favoring just the sites that attract Toolbar visitors, it's now favoring people willing to pay them a monthly fee. Way to go, Amazon. Brilliant. Now, instead of a bias against sites that don't attract Alexa Toolbar visitors, you found a way to favor sites willing to pay to be tracked. By contrast, Google has a flawed way to track site visitors, but at least they offer their tracking service for free.
Testing my hypotheses with an unranked website.
I picked a dormant domain at random that I owned for over 17 years. It had no content, so at that point it had zero traffic. I looked up the Alexa Traffic Rank, and as expected, it confirmed that it wasn't ranked. I created a page with just a few paragraphs of content, including a Google Analytics tag, then installed the Alexa Toolbar on one of my computers and visited my own site every day for a week or so. Soon, that site started to get a Traffic Rank and move up in ranks.
Shortly after that, I was surprised to see that the test site I created not only cracked the top million, but it was ranking way ahead of another site that I own that literally had 200 times more traffic. The Google Analytics code on both sites confirmed this.... yet Alexa considered the test site to be way ahead of it in their rankings.
If so many people know about the bias, what good is it?
The fact is, the Alexa Traffic Rank is actually pretty accurate... but ONLY for ranking the most popular sites, which almost everyone visits on a daily basis. Besides search engines, which often factor in a sites popularity in determining a site's position in search results, when companies try to estimate the value of a site or domain, it's Alexa Rank is often a big part of the valuation, so site owners will do whatever it takes to improve it.
The reason the system breaks down on less popular sites is because there are hundreds of millions of web sites, so even a few recorded visitors in a site ranked 1,456,789 moves it up quite a bit. In fact, a few dozen recorded visitors each day can make a site move into the desired "Top Million". Swings of hundreds of recorded visits each day easily move a site into the top 500,000 or better.
This is one reason why so many sites geared at web designers and search engine optimization companies rank well. They attract other Alexa savvy visitors, who also have the Alexa Toolbar installed, giving the impression that they have a very large number of visitors. They all know about the bias and they exploit it every way they can.
P.S. I'm aware of the bias, but I hate toolbars, but if my hypothesis is correct, this page could actually attract a lot of people who want to increase their Alexa Traffic Rank. If that's the case, even without me using the Toolbar, the very existence of this page could boost the Traffic Rank of this site, so in the interest of science, as of this writing, this site is currently ranked # 1,205,694 worldwide, and # 828,205 in the US. I will post any changes to this site from time to time to see what happens.
Joe Crescenzi, Founder
Related Media:
(Reply N/A) (Edit Topic N/A)
(Like Topic N/A) [0 ] 7863 Views
Related Posts
Amazon(30)Amazon Alexa(1)SEO(4)
Top 25 Posts
* Note: The ideas on "Idea of the Day" were posted without any formal research into existing inventions.
In some cases, patents may already exist for these ideas, in other cases, there may not be any existing patents and you are free to develop and explore the viability of developing and patenting the ideas.
The authors make no claim that any of the ideas are safe, practical, or suitable for any particular purpose. You are responsible for the results of trying, developing, patenting or using any of the ideas on this site.
For some people, our ideas are just an interesting read, but our goal is to encourage you to take action. If you see an idea that you like, do something with it... Take action.
- Joe
Idea of the Day!
on...