‘Design like you’re right. Read the data like you’re wrong.’

Before digital, designing was more of a blind process. In this essay on new White House chief data scientist DJ Patil’s previous role at Greylock, John Lilly discusses how better understanding data improves the design process. Teams used to spend “a lot of time on ethnography, need finding, doing paper prototypes and then doing basic mockups and user testing,” getting 80% of the way there before building it. Now designers can quickly build a prototype, then test and iterate based on good data (a process that can be adopted at news organizations).

+ 5 tips for evaluating assumptions: schedule regular times to evaluate, talk to people who haven’t seen your work, determine “actionable” metrics, reduce iteration time, and even if you’re big, apply the process (KISSmetrics)