Yesterday NextWeb reported on the drop in digg.com’s traffic, as did ReadWriteWeb both make for great reading – take a look. I believe they’ve missed something though.
No doubt the traffic has dropped dramatically and it is also blatantly due to the recent changes digg have made to their site. Those points are without dispute – digg managed to alienate many with their newest version.
The interpretation of Hitwise (or Comscore) data is what is flawed, even allowing for discrepancy between their estimates and actual figures.
It is certainly one interpretation to say that “Digg users are voting with their feet”, but this doesn’t take into account the fact that most people who visited digg.com in the past never logged in to the site, or at least never dugg. A user and a visitor
Consider a story which popped (hit the frontpage) on the old digg.com – for every digg there were between 10 and 100 unique visitors to the site. People used to visit the digg.com frontpage to find great content. In their droves. When the new version of digg.com was released many users simply upvoted stories auto-submitted by reddit in protest.
Imagine the effect every frontpage story being a reddit story had on people visiting the frontpage for news… The protestors had their genuine gripes, but ultimately their actions contributed in no small part to digg’s loss in traffic. Some of those people perhaps did “defect” to Reddit, but nowhere near as many as left Digg.