Digg and the Deathly Redesign (updated)

It’s a pretty safe bet that any website that goes through a “redesign” is going to end up working worse than before: it happens all the time. However, Digg has now surpassed most previous disasters, and according to Hitwise monitoring: “Since the end of August, traffic from UK Internet users to Digg has declined by 34%. In the US, which is Digg’s primary market, visits have dropped by 26%.”

In this case, of course, what Digg went through was more than the usual tasteless aesthetic Web 2.0 update: it is new design, new code, and new infrastructure. Basically, Digg threw out the whole of Digg and built a new one based on different principles. Instead of the old user-driven news, the new one would have news pumped in by publishers. Old Digg users would have buried this stuff, but they couldn’t: Digg had taken away the Bury button.

Users revolted and the new Digg was buried in hostile comments, but it had no way to go back to the previous version. Because MySQL hadn’t been up to handling the traffic, according to founder Kevin Rose, the new Digg had moved to Cassandra, a NoSQL non-relational database originally created by Facebook. (Sadly, “Facebook wannabe” runs through the new Digg like Blackpool through a stick of rock.)

In a blog post rendered almost unreadable by the clueless design, Kevin Rose responded to some of the criticisms, and some things have indeed been fixed. But a lot of things haven’t. And the Hitwise numbers so far don’t show any evidence that new users are flooding in to replace the angry defectors.

If Digg dwindles into insignificance, it won’t be the first popular web site to fail. However, it may well end up illustrating some useful points. First, if you have a user-driven site, it may be suicidal to try to take it in a direction that’s opposed to the one your “power users” want it to go. Second, it may be suicidal to redevelop your whole infrastructure and launch a new site in such a way that, in case of disaster, you can’t go back. Third and last, appearances don’t matter anything like as much as web designers think. Real users don’t go to web sites because they look pretty but because they enable them to do things they want to do.

According to Hitwise:

Unhappy with the way Digg has changed, many of the high profile power users publically rebelled by switching to another social bookmarking site, reddit.com. However, although the decrease in visits to Digg is highly noticeable, there hasn’t been a correlating spike in traffic to Reddit. During the same period of decline in the UK for Digg, Reddit only increased its visitors by 2.6%.

I’m hoping that Reddit will ultimately show a significant increase in traffic as Digg declines, because it works very well, and because it’s really not pretty at all.

Update: A post at ReadWriteWeb says: “Reddit's lead developer Christopher Slowe just contacted us with updated traffic numbers for reddit. According to these numbers - which come directly out of Google Analytics - Hitwise's numbers for Reddit are wrong. Overall, traffic to Reddit increased 24% over the last two months (mostly during the month since the Digg relaunch) and these numbers are holding steady.” Thanks to Jordan Stone for pointing this out.