“Don’t polish the cannonball!!!”

We were in the submarine attack trainer in Groton, Connecticut. It was 1995 and I was a young officer stationed on USS NARWHAL (SSN 671) a Cold War nuclear attack submarine. As part of my Command Qualification process, I was playing the role of “Approach Officer” (the Captain) during simulated battle stations at the training center. We were up against a simulated Russian Victor III or some cool sounding former Soviet submarine. I pictured myself as Captain Bart Mancuso on USS DALLAS in the Hunt For Red October. But, then reality set back in quickly at the real Captain yelled at me “Koonce, I said stop polishing the cannonball!!”

You see, my team and I were piecing together all the information we thought we needed to know where this bad guy was and where he was going so we could shoot him. Unlike an aerial dogfight, we didn’t have exact information on the position or velocity of our adversary unless we used active sonar which then would give away our position. So we had to synthesize the sounds coming from our simulated adversary into a “firing solution” so we could launch our simulated torpedo and get it close enough for the torpedo to find the enemy. We are moving through the water. He was moving through the water. The ocean currents and temperature gradients were distorting the sound. There were fishes burping and whale farts and other noise masking our adversary’s signals. There were assumptions in our calculations and accuracy issues in our equipment to deal with. Bottom line, there were a lot of unknown variables.

The Captain was “ever so subtlety” reminding me that the longer I waited to shoot the “bad guy” the more chance I give for him to shoot me first. “Lieutenant Koonce! You have to get inside his decision cycle. You have to get inside his OODA Loop. Stop polishing the cannonball!!!”

“Torpedo in the water bearing 262” – it was too late. The enemy had gotten inside my OODA Loop!

Air Force Colonel John Boyd

Lately I have been studying Colonel John Boyd and his OODA Loop. OODA stands for Observe, Orient, Decide, Act. It seems pretty simple at first. I remember being introduced to it back when I was a young naval officer learning about submarine combat tactics. When the senior officer explained it to us young pups, I probably thought something like “Well, duh!? What else would you do? Look around, figure out what is going on, and do something.” But, now that I am not as smart as I was in my 20’s, I see the power in Boyd’s model and the depth of wisdom if you study it beyond the surface (pun intended).

In my current work helping companies with High Reliability Organization principles, I am convinced that Colonel Boyd’s OODA Loop is a powerful tool to help individuals and organizations make better decisions and therefore have higher reliability in their operations.

One of the keys to understanding the OODA loop in greater depth is what Boyd called “the Big O” – Orient. The idea is that while observing, your mind is orienting what you observe into “mental models” of past experience and understanding so that you can then decide what should be done and act on it. This is a critical understanding. The world we live in is complex and ever changing. Whether it’s dog-fighting at Mach 1, a nuclear submarine attack at 8 knots, or starting up a refinery after a 45 day turnaround, the systems and processes we operate are complex and there are always changing variables and uncertainty. Our ability to observe these changing variables, synthesize them into mental models, hypothesize what action to take and then take that action while all the time observing and repeating the cycle is crucial.

In Boyd’s world it was about shooting down the other jet before they shot him. By getting inside the other guy’s OODA loop, Boyd could throw off the other guy’s plan. When you move through the OODA loop faster than your enemy, you deny them the ability to make good decisions. In a non-war setting, the “enemy” may not be so obvious. In high risk industrial operations, the enemy is the hazard that lurks just beyond the safeguards we put in place. They are constantly trying to get at you and your people.

But, don’t mistake my message. It’s not just about faster decisions. It’s about better, faster decisions. Too many people have trouble knowing when they have enough information. They polish the cannonball too long. They miss the weak signals. They delay action because they don’t think they know enough. They wait until the problems are obvious and then it’s too late.

But, what if you make a bad decision? Remember, this is a loop in a constant cycle of repeating. All the time you are making decisions you have to be observing the system’s reaction and going through the OODA loop again and again making adjustments. It’s not fire and forget. You might see things getting worse and want to stop what you are doing and place the system in a stable condition. Unlike, a fighter pilot or submarine Captain, most industrial safety processes can be shut down and placed in a stable condition. Yes, this may come at a commercial cost, but generally speaking, the cost of a safe shut down is always going to be less than the cost of a safety incident.

But, what happens when what you are observing does not match a mental model? This is where we as humans really have trouble. Our brains try to force what we are seeing into a model we can understand. We make bad decisions. If you study major industrial accidents, in most cases, decision makers tried to make the information they were getting match previous situations (mental models) that they had observed and then made decisions that did not match current reality. Part of learning about the OODA loop is recognizing when you or your team does not have the mental model to deal with the situation directly. Boyd describes this by discussing a series of “Destruction and Creation” of mental models. The better we become at observing our surroundings, “destructing” what we see into its major components and creating a new model to match our observations, the better decision makers we are as individuals and organizations. Some people are naturally better at this than others, but all can improve through practice.

I hope like me you always ask “So what?” Well, here are three things you can do as a leader to improve decision making and reliability in your organization:

Build Mental Models: The reason I was in that trainer back in 1995 and my Captain was “guiding me” was to build my mental models. That’s what training does. Education is about gaining knowledge. Training is learning how to apply the knowledge (behaviors) and thus build mental models of how to deal with complex situations. Training can include expensive simulators but can also be very inexpensive “table top” exercises. Every week on a submarine we would hold supervisor “table top” training where we worked through interactive scenarios using nothing but a dry erase board and the procedures to discuss complex scenarios about the nuclear propulsion plant. These one hour training sessions were powerful to build mental models for complex operations. You can do the same at your company. Imagine if United Airlines had simple training like this about what to do when a passenger refuses to give up his seat. I doubt letting airport police drag his bloody body off the plane with video cameras rolling would have been the outcome of that training exercise.

Encourage an observation culture: Management By Walking Around has been talked about a lot, but the real benefit comes when the person knows that they are there to observe operations with a purpose – as part of their OODA loop. It’s not just a conversation with the team to make sure they know you care. It’s matching your observations of the “work as performed against the work as planned.” This deviation of work planned versus work performed is finally being recognized as one of the issues in Operational Excellence and Human Performance. There are many ways to get “eyes on target”. Observations of maintenance, control room operations, training, drills conducted, project planning sessions, shift turn overs, and audits to name a few. But, you have to be intentional and have a purpose. You have to set clear expectations and, if needed, train the supervisors, managers, and executives on how to conduct an effective observation. And, please, don’t let a senior executive tell you he or she can’t add value because they don’t know the systems and operations. Sometimes, the most revealing information about operations comes when a person asks an innocent question about a system they know nothing about. I used to ask a “dumb” question like this: “Pretend I am your Grandma. Explain how this system works to me.”

Create Feedback Loops: Organizations have OODA loops just like individuals and OODA loops work at the strategic as well as tactical level. One of the challenges we have observed is that many organizations don’t have good methods for feedback on operations. Unless something really bad goes down, most civilian organizations just move on after an operation to the next project or operation. In the Navy, we did “Hot Washes” after successful events to capture what we did well and any little things we could do better. There were always ways to get better. After every Blue Angel performance, the Angels spend hours going over the entire flight. You have to match the investment of your organization’s time to the return.

Well, that’s enough polishing of this cannonball. I am going to launch this and see if I hit the target. Where is my mental model of OODA Loops wrong? You see, even when discussing OODA Loops, you have to use OODA Loops.