Nov 15 2007
Measuring Website Traffic – Alexa vs Compete
External measurement estimation of website traffic usually evokes controversy. I thought I’d have a look at the two major providers of (free) traffic ranking data.
Alexa rank is often used to compare websites, and is probably the system referred to most frequently by bloggers, and in creating “top lists”.
Of course there are many criticisms levelled at Alexa, including;
- data is collected solely by the installation of the toolbar,*
- many anti-virus programs detect Alexa as “suspect” or “malicious”,
- this may lead to sites that are more tech oriented gaining an advantage (even more so with the release of an Alexa sanctioned FireFox toolbar called Sparky),
- popularity amongst Chinese users, which gives Chinese websites and advantage
* although some assert that “Alexa redirects” help boost Alexa ranking.
In saying that, however, Alexa claims over 10 million toolbars have been downloaded, and they publish rankings on a country level which is extremely useful information.
Simon Chen, who has has some excellent coverage of Blogworld Expo, has this to say about Compete:
3. Alexa rankings are considered by most of the web savvy folks as useless. I agree. Compete is much better due to the fact that they pull data from ISP’s, their toolbar and various opt-in panels (as opposed to Alexa, who depend only people installing the Alexa toolbar).
So on top of 2 million plus toolbar users, they also get real data from ISPs, ASPs and opt-in panels. This data is then “aggregated, transformed, enhanced and normalized” by their team of mathematicians, statisticians, and data scientists. Phew! No wonder it appears so much more credible.
While this all sounds very promising, there’s just one problem. It only counts U.S. visitors. Even if you have a toolbar (which I’ve had for a while) – if you live in another country, the information is completely disregarded, at this stage. It appears that “international usage calculations are in development”.
This is certainly evident in the graph below which compares this blog to dLook. Blogpond (over the last 6 months) has received 38% of visitors from the U.S.
dLook, which is obviously highly targetted to an Australian audience, receives less than 2% of traffic from the U.S. But in terms of overall traffic, dLook receives many times more than this blog.
It appears at this stage, that Compete provides little value for Australian websites (targetting an Australian audience), but could be of interest if they expand their data collection to our shores.
Other services that are used in Australia are Hitwise and Nielsen//NetRatings, but little is publicised about their methodology, and of course you have to pay a premium to get access to their data.
Next I’m going to look at metrics for the Top 100 Australian Blogs Index.
13 Responses to “Measuring Website Traffic – Alexa vs Compete”
Compete looks interesting, I have never heard of it. BTW when is the next top100 update due?
Hi Thomas
The list should be up later today (probably tonight), with my fortnightly update post tomorrow.
As you well know, Alexa drives me nuts!
I have been watching compete for a few months, even with the lack of nonUS stats they seem more acurate than Alexa. If they could cover everywhere then it would be a fantastic service, but I guess that isn’t likely to happen anytime soon.
There is http://www.webranking.com.au for Australian sites, which also says it get its information from ISP.
I am glad that I came across your site Meg. I am learning so much! I have been totally unaware of any of these services. As a newbie blogger I just rely on my wordpress stats counter to give me some indication of where people are coming from (and then it is just from a direct ‘click on a link’) and how many people are visiting to see if what I am writing is entertaining or interesting to others.
So besides increasing your blog rank what is the benefit? And do these things increase spam?
Hi Leigh – Compete does look to be reasonably accurate in representing US visitors. I’m not convinced about webranking.com.au, which is why I didn’t include it. They don’t mention which ISPs & do say that there can be some bias. And besides they don’t rank me so what good is that? 😉
Hi Kelley,
(Ditto about your blog!) Both are really just tools to rank websites, or estimate the traffic. The Compete toolbar probably wouldn’t be worthwhile downloading at this stage, but I’d certainly recommend alexa. Some people get concerned about privacy, because this toolbar does track your movements over the web, but Google does similar every time you search, so it doesn’t bother me.
I haven’t noticed any increased spam. You don’t have to register for Alexa, though they do ask you to (optionally) provide a bit of non identifying demographic information after you download the toolbar.
Meg, I had been following it for about 12mths wondering if AFW should be in there, so I emailed them and put me in. dlook is there though.
But if they maintain they rank the top 10,000 websites, shouldn’t it have been there without you having to ask for it? Maybe I just don’t make the top 10,000.
For my 5 cents worth i consider Alexa to be next to useless as it isn’t a true measure of website popularity at all but rather a measure of popularity for people that happen to have the toolbar installed. While i am not sure of the percentage of the whole number of internet users i would be suprised if it was even 10%
So saying for the price you pay and the information you get it is value for money.
Interestingly enough a lot of advertisers rely on the information form alexa when making their decisions as to who to advertise with as it can produce skewed and erronous data that i wouldn’t trust although i do still accept their advertising
Ou… it’s the first time that i read about Compete…. So thank you for writing about it. It’s seem to be a very good service.
Best regards, Riccardo Giuntoli.
@Greg: You fail at statistics.
@ Riccardo – glad you found it useful
@ Bob – just wondering why you say that?
Is there any way to get traffic data for even small websites? All I can get on Complete, and I’ve seen on Alexa, is for the top websites (which is naturally only a small percentage of all the websites on the Internet).
Hi David
I’m not sure exactly what you are asking, but if you go to Compete or Alexa you can query any website’s traffic. Alexa ranks around 10 million websites, but doesn’t really tell you that it gets x number of visitors, just where it ranks. As I understand it Comptete will estimate traffic for lower volume sites.