Many moons ago, I interviewed at Microsoft for a Program Manager (PM) intern position. I was not offered a position but it was an enlightening experience nonetheless. The process consisted of several interviews with various members of a project team, and they asked a mixture of design and programming questions. Most of the programming problems were pretty straight-forward, string manipulations and such. The most of the design questions were more difficult, but still do-able. Most were related to programs I was relatively familiar with, like IDEs, email clients, browsers, text editors, etc. But there was one question in particular that stood out:

“How would you design a gun for visually impaired people?”

Well, now that is an interesting question. I had prepared for the expected “how would you design a better {fruit peeler, telephone, elevator}?” questions, but guns for blind people? I was speechless for a few long seconds.

During that brief silence, my brain was racing through everything I knew about guns and trying to reconcile that with what I thought PMs were supposed to do. I knew that PMs were responsible for ensuring that the product met customer needs, so I had to learn more about the user requirements. Is the gun for self defense or hunting? Is the user totally blind, or severely myopic? Are there any physical concerns to account for (like weight, recoil, handedness)?

I imagined myself blind, and then thought about what I would want in a gun. I’d want to be able to safety-check it quickly. I’d need to be able to load it and clean it easily. Ideally, malfunctions would be rare and easy to clear. If I wasn’t totally blind, I’d want simple and highly visible sights. After much deliberation, I had narrowed it down to a large frame revolver or a shotgun. I was thinking that a revolver like a Ruger GP-100 (with Hogue combat grips) would be easy to grip, you could easily check if the chambers were empty or not, has no safeties, and is very reliable. A pump-action shotgun would be reliable, easy to maintain, offers multiple chances to hit a given target per shot, and has dead-simple sights. 12 gauge is a versatile, proven system. With the right barrel, choke, and ammo, it is good for everything from deer to doves to bedroom intruders. Eventually, I settled on a Remington 870 Wingmaster in 12 gauge with tritium night sights and 26 inch barrel.

This was not the right answer. The interviewer seemed a bit confused with the detail of my response, and I was confused that he was confused. (That should have been my first clue.) I did what I thought the PM was supposed to do: analyze requirements, empathize with users, and design solutions. I felt that I had met the requirements; I had created a weapon system for blind people. I did (almost) everything right, and was still wrong. Finally, I asked the interviewer what the correct answer was. The interviewer smiled gently and said

“Should we really be doing this? How many people are going to buy this? Is it even ethical to do this?”

I realized I was too focused on the how of the task, and not the why. The PM has to care about both. In my case, I knew what problem I was trying to solve (blind people can’t shoot) and how I would solve it. I forgot to ask the follow up question of “why are we trying to solve this problem?”.

At the time, it seemed like a trick question. In the real world though, upper management is eventually going to hand you a stupid project. The PM is the only thing standing between you and that black hole of time, money, and energy. And it’s up to your PM to say “should we even be doing this?”

P.S. In my defense, it is legal for blind people to hunt in Michigan and Texas.