The perfect is the enemy of the deadline.
Bill Wallace, engineering group manager for GM’s Volt battery. Photo by Steve Fecht for GM.
When we had to build the hardware for our fourth major design iteration of the Chevy Volt battery, we were working closely with our suppliers and realized we needed a three-month delay to do it. They said, “That’s inconceivable. You can’t do a whole redesign in three months.” There were a lot of shocked faces in the room when we said we weren’t going to stop. What do you do? One, you work harder. I didn’t sleep a lot — that’s the obvious part. Two, you scale back expectations. You’re better off doing 95 percent on time than 100 percent late. When you’re late, nothing happens: The battery isn’t in the car and you’re not learning anything. We made the date with about 98 percent complete.
— Bill Wallace, engineering group manager for GM’s Volt battery
Always know whose problem you’re solving.
Lisa Watts, Nokia’s director of corporate business development.
— Lisa Waits, Nokia’s director of corporate business development
Make sure you’re talking to the right people.
Josh Handy, head of industrial design at Method.
In April 2007, we launched Bloq [a line of body washes and lotions]. That product was very difficult for us. It didn’t make us any money. We were growing extremely fast at the time, from 30 to 90 people in a year. The formula inside the bottles was much too thick to be dispensed with a bottle that was hard to squeeze. We knew what the bottle was like, but we didn’t test it with the formula. When it finally came together, it was too late to change. That sort of oversight is a ridiculous mistake. The right people weren’t talking together enough. People often get into silos in their heads, so we tend to move them around the office every six months. It drives them nuts, but it allows them to build deeper connections in the organization. Now I sit right next to the formula guy.
— Josh Handy, head of industrial design at Method
Ask yourself, “Is this necessary?”
Peter Deng, product manager at Facebook.
We have ideas on how we can help users communicate better, but we’re not always sure how well they will work. We do tons of testing for every little change to the page. One thing I know now, which I wish I had learned earlier in my career, is how to prevent “feature creep.” You have to be able to recognize when you’re inventing a feature that duplicates another or only adds unnecessary complexity. Often we’ll draw a decision tree, mapping out users A, B, and C to figure out how much use a feature will get. If something doesn’t affect A or B at all and only affects C in a few circumstances, that shifts your priorities pretty quickly.
— Peter Deng, product manager at Facebook
