I’ve been in the field of software engineering for a decade now. I’ve had the chance to work with many leaders, and to be one myself. This post contains some anti-patterns I’ve seen and at times fallen into in my roles.
At MIT, in the Mechanical Engineering Department, Course 2.003 is Dynamics and Controls. Students in this class, at least when I studied there, learned how to model and simulate the motion of complex systems. The final project in the class was to design a control system to keep an inverted double pendulum upright and stable. The control system was hooked up to an actual linear motor and pendulum. Instructors could load each student’s program onto the machine and run it against this real system. Students could test out their program by bumping and poking the pendulum, and watching it stabilize. To see a system like this in action, watch this video.
Inverted pendulums are fun systems to play with. You can try one out yourself by balancing a broomstick on your hand. Changing the length and weight of the arms can make it easier or harder to control, but with enough practice most people can get it balanced. The Exploratorium in San Francisco even has a double pendulum you can try to balance yourself — I haven’t seen anyone get very far. There are some videos on Youtube of triple pendulums and there is a research paper from 2004 discussing a theoretical model for a quadruple pendulum, but I haven’t been able to find any videos.
Management as a Balancing Act
Many managers and team leaders fall into running their teams like they would balance an inverted pendulum. These leaders have a vision of where the team or project is supposed to be going. Each day, they come into work and check on these team members and projects to make sure everything is still going in the right direction. If something is starting to go off track, they gently (or not) steer it back on course, just like they’re balancing an inverted pendulum.
This is a rewarding style of leadership! The manager gets to jump from issue to issue, putting out fires and solving problems. The results are immediate, the gratification instant. Something that was off track is now back on track. This success leads to rewards. More reports to manage, more projects to direct.
Naturally, some of the team members will step up, and start to take on leadership roles of their own. With reports to help put out the smaller fires, the manager can focus on the bigger, larger ones. After enough time and luck, some of these leads might even mentor their own leads.
The organization is growing very well. Until it stops. Team and group productivity slows down. The teams and individuals aren’t working together as well as they were before. People are upset and stressed out. In some cases, it might even feel like they’re working against each other. It doesn’t make any sense! There are more fires popping up than the manager can possibly deal with. Still, they try, and eventually burn out — unable to diagnose the problem. A reorg happens, teams get shuffled, someone moves on to a new adventure.
What Just Happened?
How do we understand the root cause, and prevent it from happening again? We have to go back to our pendulum model. Each project a leader is overseeing, each report the manager is overseeing, is a balancing act. An inverted pendulum. Maybe an inverted double or triple pendulum. But gravity is faster than enterprise companies, so the failures take longer to notice.
With practice, a manager can handle a bunch of these at the same time. One day a week per project, maybe half a day, is all it takes to keep things on track. Check that the project is going in the right direction, nudge it back into place if it’s not. Move onto the next one. Check that each report understands their role in the organization and is happy. Give them some guidance and direction if not.
Very good managers can even keep complex projects — double pendulums — balanced temporarily. Like backing up a truck with a trailer, the fix might be counter-intuitive, but the manager is smart and experienced. They identify the right direction to push their pendulum, which corrects the second and keeps things on track.
Until there are too many of these pendulums. Or things get too complex and turn into triple or quadruple pendulums. So what do we do? Quadruple pendulums are impossible, yet large companies everywhere are comprised of systems 10 or 15 layers deep.
One of the most obvious solutions is train and mentor excellent leaders inside the organization. Teams are not exactly the same as an inverted pendulum — leaders within the organization can act as controllers on their own. This is a great idea, but doesn’t work forever. A team operating like this might stay oriented correctly for months, or even years, especially if the leaders and managers embedded within the pendulum are very good and aware of what is going on.
But even the best leader stuck inside a complex quadruple pendulum will eventually be unable to keep the entire thing upright — there is not enough leverage from their point in the system and communication is too hard for them to see the big picture. A large enough perturbation will eventually cause the whole thing to come crashing down and need to be rebuilt from the ground up.
The problem is that these systems are inherently unstable. The mental model of a manager in this organization is that the world is falling down around them, and they need to keep things running correctly. The leader needs to make the problem stable. They should try to build a system that remains in place without their interaction. Great leaders strive to make themselves obsolete.
Jocko Willink calls this technique “decentralized command”. Most other management books and guides call it empowerment and delegation. The overall idea is to push as much decision making down into the organization as possible, where the employees have the most information. This is a great idea, but only works to a point. Leaders need information to make the right decisions, and Metcalfe’s Law makes communication hard and information propagation slow in a large organization.
Flip The Pendulum
The solution should be obvious here. An inverted pendulum is unstable. A normal pendulum is stable. Flip the model upside down. Anyone can keep a pendulum 10 layers deep correctly oriented. Gravity does the work for them.
So how does a manager do this in their organization? Communication and direction setting. Leaders should strive to over-communicate everything. Over-communicate the mission. Over-communicate the values. Over-communicate the principles. Repeat this until you think there is no possibility of a misunderstanding, then do it one more time for good measure.
Quizzes are generally bad, especially pop quizzes, but this is a time when they can be useful. Ask your reports what they think the organizational priorities are. Ask them if they know what is expected of them. Ask them if they know what they should be doing now, in a month and in three months. Ask their reports the same questions. Make it clear though, you are not evaluating them in this exercise. You are evaluating yourself — specifically the clarity of your vision and your ability to communicate it. It is not your employee’s fault for getting any of these answers wrong — it is your fault. Again, it is not your employee’s fault for not understanding the team’s mission, goals or values. It is your fault.
Everyone has different communication styles — figure out what works best for you. In general though, written is better than verbal, and broad is better than one on one. Feel free to announce changes or goals in all-hands meetings or reinforce them in one on one meetings. But also communicate everything in writing. The act of writing forces you to clarify your thoughts in a way that most people cannot do purely in their heads.
Broad, written communication eliminates the telephone game effect by guaranteeing everyone receives the same message. You can revise and edit it in a way you can’t do verbally. You can share your message and test it by quizzing a small set of readers, and repeat this until you’re sure it’s clear. When your entire organization has the same clear vision in their minds, work feels easy. Problems resolve themselves before becoming large issues that require your attention, because each team member understands what is expected of them and how they are allowed to accomplish it.
In a non-inverted pendulum model, work shifts from day-to-day course correction and putting out fires to communication and clarification. This is less immediately gratifying, but don’t confuse that for less important or effective. If everyone knows what to do and how to do it, a leader’s job is done.
That’s no excuse for boredom (or golfing) though! Goals and plans change, and their communication takes time. You’ll likely find that the world you’re operating in is changing as fast or faster than you’re able to communicate these changes. And that is your new mission! Keep everyone as informed as possible in an ever-changing world! Refine the organization and teams to facilitate faster communication and reaction times. Continuously improve yourself and your team. The direction and vision don’t have to come from the leader. Many organizations do very well with bottom up leadership — where the team members closest to the problems are responsible for deciding on goals and strategy.
Don’t make the mistake of confusing bottom up leadership with turning your team into a fully self-driving organization. In this environment, the leader’s role is to ratify good decisions and communicate them. Direction can be set elsewhere, by anyone, but a manager is still in the best position to cement good decisions and make sure everyone is on the same page.
Do you need to flip any pendulums in your work life today? Here are some signs you might:
- Is the majority of your time spent dealing with urgent issues?
- Is your organization struggling to “stay on the same page”?
- Are leaders in your organization struggling to step up?
- Are you the only one capable of fixing urgent problems?
- Is the majority of your communication through one on ones, direct chat messages or emails?
If some or many of these are true, think about the pendulums. The solution isn’t to get better at balancing them — that will only work for a small time. The solution is to work with gravity, not against it. Let urgent problems fall through, giving team members the opportunity to step up and fix them. Spend your time clarifying and communicating direction in a group setting.
What do you think? I’d love to hear feedback here or on Twitter.
Special thanks to Kim Lewandowski and Rajeev Dayal for reading drafts of this.