Thursday, October 9, 2008

Social media and user feedback

Here's something you would have never seen prior to social media - users banding together to critique your product, using your product.

This month Facebook redesigned their interface, with the aim of reducing the clutter, and a number of users joined in protest by forming Facebook Redesign Protest groups.

The redesign was well executed, but what really impressed me was the response to the protest by Facebook execs. They are very confident that users will love the new redesign, and they find it humerous that people are using their product to protest their product.

The immediate feedback loop of Web 2.0 can be your downfall if you're not careful. Some Web 2.0 companies only look at feedback, and things like split-testing (creating two pages with two different designs, and serving each to half of your users to see which gets a better response rate or better feedback). In chatting with one of our Sr. Usability Architects, Scott Plewes, last night, he related that any time you don't use direct observation as part of the user-feedback mix, you risk missing the context and misinterpreting the data. Bottom line - you need to understand the motivation behind a user's decisions or you are flying blind.

I don't have the inside line on the Facebook redesign, but I'm sure Facebook employed a number of testing methods to gather feedback on the design - things like focus groups, ethnographic research (observing users using Facebook in their natural habitat), and usability testing. Where a number of Web 2.0 companies would have hit the panic button and rolled back the changes when they saw the protest, Facebook execs were calm and confident that they did the right thing, no doubt because they did their homework.

1 comment:

Oren Mazor said...

I found it very interesting that by default every time you loaded the facebook page, it would push you to the new design, but would provide you with a link to go back to the old one.

This gave both the user a way to compare the two designs, and the establishment with a really easy way to monitor which use cases result in the user switching from one design to the other.