People love sticking numbers on things – usually in order to compare and contrast them with similar things. Websites are no exception: after all, what webmaster wouldn’t want to compare and contrast his or her website with that of a competitor? One of the most well-known metrics aimed at allowing people to do exactly this is the infamous Alexa Rank – why infamous? Because historically heaps of people tend not to trust it as an accurate indicator of what it is supposed to be a measure of, namely: ‘how a website is doing relative to all other sites on the web over the past three months’ (quoted directly from Alexa’s official website)* – which is itself a bit of a dubious definition: I mean ‘how a website is doing’… What does that mean?! Right, enough of an intro, let’s get stuck in…
A little history
Founded in 1996, Alexa is a California-based subsidiary company of Amazon.com (acquired by Amazon in 1999) that specializes in providing commercial web traffic data gathered via various toolbars and web browser extensions. Some of Alexa’s most notable previous activities include providing a database that served as the basis for the creation of the Wayback Machine and the creation of various search facilities (now largely discontinued). However, the thing they’re probably best known for is, of course, their ‘Alexa Rank’ – a metric that ranks websites in order of popularity or ‘how [well] a website is doing’ over the last 3 months.
How are Alexa Ranks measured?
According to the official Alexa website’s Our Data page, the rank is calculated using a ‘combination’ of the estimated average daily unique visitors to the site and the estimated number of pageviews on the site over the past 3 months – with the site with the highest combination of unique visitors and pageviews being ranked as #1. The data is collected from a subset of internet users using one of 25,000 browser extensions for either Google Chrome, Firefox, and/or Internet Explorer. An algorithm then ‘corrects’ for various potential biases and attempts to compensate for visitors who might not be in Alexa’s measurement panel (a factor it historically hasn’t always tried to accommodate for) and normalizes the data based on the geographical location of visitors.
How can I view a website’s Alexa Rank?
At least this part is straightforward: simply go to the official Alexa website, type in your full domain name and hit return! What’s more, scrolling down the results page reveals no end of other interesting metrics, such as Bounce Rate, Daily Pageviews per Visitor, Daily Time on Site and the Percentage of visits from Search, as well as various Demographics, a list of sites that link and even page speed/load times! All of which should probably be taken with a pinch of salt…
What does the Alexa Rank mean?
The general consensus seems to be a reluctant admittance by most (certainly not all) that there does indeed appear to be a very general – rough – correlation (seemingly with a LOT of outlying data points) between a site’s Alexa Rank and traffic for well-established websites that receive over and above a certain level of traffic: i.e. for relatively popular sites with Alexa Ranks of less than somewhere in the region of about 50,000 to 100,000 – to their credit, Alexa does actually state on their website that ‘traffic rankings of 100,000 and above should be regarded as [very?] rough estimates‘ and that conversely ‘the closer a site gets to #1, the more accurate traffic rankings become‘. Anything higher than this 100,000 waypoint and everyone I’ve ever spoken to on the subject (I hangout in somewhat geeky circles) seems to pretty much laugh it off as a metric that can, nine times out of ten, be almost completely ignored.
Why don’t people tend to trust it?
It seems that a lot of people/webmasters with access to reliable web analytics data (the most popular being provided by Google Analytics of course) for more than one website often report seeing web-traffic trends and statistics that appear to be completely out of line with the corresponding Alexa Rank for each site. Unfortunately, you’ve only got to Google something like ‘Is Alexa Rank Accurate?’ to find such reports – not good! Although, in all fairness, quite a few of these reports are for websites with quite high Alexa Ranks – i.e. often way in excess of the 100,000 mark.
The real kicker though, seems to be two-fold: 1) Alexa aren’t able to gather their data from everybody (they obviously simply don’t have access to everybody’s browsing habits) – they can only gather data from a subset of a few million users via certain browser extensions (as previously mentioned) – something common sense suggests will almost certainly skew the data right from the get-go (since the average user presumably arguably isn’t likely to have installed any such browser extensions) and 2) rather than address such concerns head on and be completely open about exactly how this particular problem is accounted for, i.e. by being more open about exactly how the underlying data is collected and used to calculate the rank, Alexa seem to (as far as my admittedly somewhat limited research goes) be less than 100% transparent on the matter, simply stating that Alexa Rank ‘is calculated using a combination of the estimated average daily unique visitors to the site and the estimated number of pageviews on the site over the past 3 months’. Hmmmm….
Are Alexa Ranks important?
For most site owners, ‘how a website is doing’ is of course very important, however, when assessing your own website my advice would be to simply stick with Google Analytics data rather than to go attributing any significant meaning to your site’s Alexa Rank. When looking at competitor’s sites however, by all means take a quick peek at their Alexa Rank for a very rough idea of how popular their website is relative to yours (assuming the same kinds of people visit both sites – thereby hopefully minimizing some of the biases brought about by the significantly-less-than-perfect way in which Alexa gather their data); however I definitely wouldn’t go thinking a particular website gets more traffic than another merely on the basis that its Alexa Rank happens to be only a few thousand lower – and if the website you’re interested in happens to have a rank of anything even near the aforementioned 100,000 mark, it’s probably best not to go attributing any significant meaning to comparing Alexa Ranks at all!
I personally hope that Alexa continues to work towards really getting this particular metric up to scratch… because it sure would be nice to stick a number on each and every website that would allow us to compare and contrast them all with a decent level of accuracy! Unfortunately, it would appear that they’ve still got quite a long way to go in order to regain the industry’s faith in the matter!
What do you think? Have you noticed Alexa Ranks becoming more accurate as they decrease? On a scale of 1 to 10, how much faith do you generally place in the accuracy of a site’s Alexa Rank?