How Am I Doing?
Ranking lists have always been popular. The Billboard Top 100, The New York Times Best-Sellers Lists, and the Dave Letterman Top 10 List are just a few examples. It’s no surprise that after the Web took off in the mid-90’s, Web popularity tools came along that claimed to rank websites by traffic.
The most popular service was Alexa, which gathered data by offering a toolbar for Internet Explorer. Released in 1997, this toolbar offered functionality missing in early versions of IE, and was installed by hundreds of thousands of people across the Internet. In exchange for the extra features, the software tracked websites visited by the users and reported this information back to Alexa.
The quality of their data has always been criticized – just do a Google search on “alexa accuracy” for hundreds of articles on this topic. The primary problem is with low-traffic sites, which are the majority of sites out there. There’s just not enough sample data coming in for an accurate ranking.
Back in 2006 I decided to research these services in more detail. Using the hospital websites listed in U.S. News and World Report as a starting point, I tracked the rankings for each hospital across all three services. There’s still a noise issue. The top 20 hospital sites were pretty consistent across all services, but it gets more jumbled once you get down to sites with less than 100,000 monthly visitors.
Despite this, one service stood out. Quantcast.com had accurate numbers for all of my websites, even ones with only a few thousand visitors a month (See this example). Their data gathering combines panel measurement, along with raw ISP traffic logs from major service providers. Their full methodology is detailed here (caution – lots of buzzwords)
Spot checks with the Web managers of a few dozen sites on the list confirmed that the Quantcast numbers were close – usually within 25% of the actual number. They also offer additional demographic data on site visitors, which makes it a good market research tool.
I continue to track hospital site rankings, and have expanded the list to include some of the large commercial and government medical sites. Now, however, the data comes from Quantcast alone.
Quantcast data was checked on Feb 8, 2009, so it can be seen a snapshot of site activity. Obviously. the numbers change all the time, but over the past three years I’ve found most stay at a consistent level of traffic. The spreadsheet has data on over 370 organizations. It also has links to the hospitals YouTube, Facebook and Twitter sites.
I welcome your feedback and comments. Let me know if you find this useful.
(note – this is an update of my article posted on webiscope.com in August, 2007)