But as a fighter pilot, I knew the essence of the problem. These guys could not land a project. We learned as a fighter pilot that we had to very carefully bang that airplane right on the end of the runway. And if we didn’t, we might go halfway down the runway and slide off the runway into the trees. And that’s what they were doing every project — sliding off the runway into the trees.
So I went to the CEO and I said, “This bank is totally screwed up. If you give me the worst business unit in the bank, I will fix it, just like I fixed Company L2”.
And he said, “Sutherland, if you want that headache, you’ve got it”.
So I said, “OK. I will report to you once a month to you and the senior management and the rest of the time you stay out of my unit. We’re going to run this as a little company in a company like a startup”. And so we broke them down into small teams, sales, marketing, installs, engineering, everyone — all working together with team incentives in a collaborative space. And we ran weekly cycles. We began building a backlog of what we needed to do. And I began to show them how to land the airplane.
How is it that a pilot can make a perfect touchdown? He has to look at altitude, airspeed, rate of descent, the heading of the airplane, understand the wind and the weather, and every few seconds being adjusting constantly. So every week they would try to land the airplane at the end of the runway. Bang! And week after week they did it, and surprise, surprise, in less than six months, that team was the best team in the bank. They had gone from the worse money losing unit to the most profitable units in the bank. Because they made their work visible. The team was given the responsibility to fix the problem and they self-organized to make it happen. It’s all about learning how to land the airplane.
Now from that banking experience, I was asked by a series of companies to come in and deal with tough problems. So I needed to figure out how to do this consistently and get other people to do it other than me. I couldn’t be there all the time for everyone. And so I began to think about how would I do this. I was running a little French company in Cambridge, Massachusetts on the MIT Campus. And five graduate students from MIT came by and they said, “We’re starting up a company building robots, can you rent us some space?”
And I said, “Sure. We have a couple of extra rooms you can use for a lab”.
And they were building these insect-like robots. And every other day the robots would come wandering into my office. They were trying to hunt me down with heat seeking sensors. And a senior professor at MIT was a co-founder and he would come by on Fridays and he would see how they were doing it. I asked him, I said, “Professor Brooks, how do these robots work?”
He said, “Jeff, the first thing you need to understand is that for thirty years, we’ve tried to build a smart system at MIT. And it’s been a total failure. We tried to build this command and control system with really big computers and huge databases. And it didn’t work. The best it could do was a smart chess program. So we said, I watched insects and I wanted to create something that had distributed intelligence, a chip in the leg that can move a leg. A chip in the spine that can coordinate legs. A neural network in the head that figures out where to go”.
And he turned it on and the legs flapped and waddled to its feet, it was like a baby learning how to walk, in three minutes it was running around the room.
I said, “Wow, that’s amazing. I used to have these really slow developers. Maybe we could give them some simple rules. And they would learn how to boot up into a super intelligent team, do you think that’d work?”
And he said, “Why don’t you try it?”
Well, I did a short time later. A company — a very successful company hired me in to build a new product. It was going to replace all their old products in half the time they’d ever done it before. And I had to figure it out. And our product had to be used by really big companies and I had to tell them how to use the product in an agile way. And we studied the literature and we found the best paper in the Harvard Business Review, where Nonaka and Takeuchi, two Japanese professors had shown three different styles of leadership. Number one, the Gantt chart. You know what happens with that one. Number two, a transitional strategy at Fuji Xerox. And number three that what they saw in the best manufacturing plants in the world.
So we ran how to implement number three and Takeuchi and Nonaka said, it needs to be self organizing, self-motivated teams. Management needs to let go and step back, they need to get out of the way. So the teams can figure out what to do.
So we implemented that in 1993. By 1995, I got together with my partner Ken Schwaber. We started rolling it out to industry. In 2001 we wrote the Agile Manifesto. And since then it’s gone everywhere in the world, all the major software companies. They’re building fighters, agricultural machinery, space probes. [Joe Justin] in the picture there is showing a car company, how to build a car in one week cycles. A new car every week.