Digg looked great. Even original digg looked way nicer than current reddit. The problem was big redesign.
They really fucked it up. Ads were being shoved left and right without knowing what really was an ad and what wasn’t. It’s the same fuckery reddit started doing and scaled a bit back. I still hate current reddit (all the pop ups are really annoying)
Moral of the story: don’t do massive redesigns. Listen to your customers (not your VCs). Change one small thing at a time, A/B test it, UX interview it with customers, then roll it more % of customers. if no red flags, increase % over time to 100%.
The classic example of a redesign done right is with eBay.
eBay's background used to be a (very) bright yellow, which looked pretty bad even for the web in 2009. When they initially changed it to the current white, the designers received enough complaints to change it back for fear of loosing parts of their userbase. They still wanted to change the background to pave the way for other websites, so slowly, over several months, they lightened the shade of yellow until it became white. [a]
This concept has been applied by many other "big" websites - Amazon, Dell, Google, and Yahoo come up as examples. [b]
It seems like any time a company backtracks on a design decision that they simply didn’t do their research thoroughly, if at all. All too often web pages look and feel is left to the software people who make terrible changes. Companies fail to realize that the webpage is their customers’ window to their business. Why would you leave that in the hands of coders?
Comments
Digg looked great. Even original digg looked way nicer than current reddit. The problem was big redesign.
They really fucked it up. Ads were being shoved left and right without knowing what really was an ad and what wasn’t. It’s the same fuckery reddit started doing and scaled a bit back. I still hate current reddit (all the pop ups are really annoying)
Moral of the story: don’t do massive redesigns. Listen to your customers (not your VCs). Change one small thing at a time, A/B test it, UX interview it with customers, then roll it more % of customers. if no red flags, increase % over time to 100%.
Oh man, I really dislike the new reddit. It's slower and wastes half the screen real estate. Why did they think it was good?
Seriously a bummer, as you know the old design is gonna be phased out at one point or another.
The classic example of a redesign done right is with eBay.
eBay's background used to be a (very) bright yellow, which looked pretty bad even for the web in 2009. When they initially changed it to the current white, the designers received enough complaints to change it back for fear of loosing parts of their userbase. They still wanted to change the background to pave the way for other websites, so slowly, over several months, they lightened the shade of yellow until it became white. [a]
This concept has been applied by many other "big" websites - Amazon, Dell, Google, and Yahoo come up as examples. [b]
[a]: https://theuxblog.com/blog/redesigning-user-experience
[b]: https://articles.uie.com/death_of_relaunch/
It seems like any time a company backtracks on a design decision that they simply didn’t do their research thoroughly, if at all. All too often web pages look and feel is left to the software people who make terrible changes. Companies fail to realize that the webpage is their customers’ window to their business. Why would you leave that in the hands of coders?