As reported in the NY Times, the debate continues about measuring web properties. It was perfect timing for me because, I was chatting with Robert Scoble at Web 2.0 about the size of the feed reader audience. Unbeknownest to him, I obsessively attempt to track the core feed reader segment (Bloglines and Google Reader.)
The NY Times article surprised me because, I didn’t think that the debate continue for top tier web sites. Surely, they are big enough and don’t have large discrepancies. I thought, most of the panel based approaches are fairly good at measuring at audience size. But I guess I’m wrong. For example, Forbes.com claimed 11.6 million United States visitors last month. Yet Nielsen/Netratings only counted 7.5 million and Comscore estimates were even lower, 5.8 million. The case gets even worse for smaller sites which can’t rely upon third-party measurement.
The problem is an old one for sites without registration. These sites track unique browsers which is some major flaws. The first is that I might use multiple browsers (PC and Phone. Or Laptop at work, Desktop at home, Phone and iPhone.) The second is that I might upgrade my browser which would count as two unique browsers. In short, the tracking of unique browsers needs to get collapsed to unique users. Withou a required unique user id, it’s difficult to be sure that a specific user is using a specific set of browsers.
Over the next couple of weeks, I’ll write a series of posts outlining the different data available from the various services and outline some of the caveats. I think this is important because most of the tech bloggers are comprised of independent developer types who generally don’t have access to paid subscription data services and therefore have probably not delved into the details of the services.
