In the summer of 2002, the U.S. military ran a war game called Millennium Challenge. It cost about $250 million. The enemy was a hostile country in the Persian Gulf, and the exercise was meant to test the ideas the American military would fight with in the years ahead.

To play that enemy, the organizers chose Paul Van Riper, a retired Marine Corps lieutenant general. He knew the American side would rely on its surveillance and electronic networks to see everything he did, so he gave them very little to see. He sent orders by motorcycle courier and launched aircraft using light signals instead of radio. Then he struck first, with cruise missiles and swarms of small boats.

The opening strike sank at least 16 American warships, including an aircraft carrier.

The organizers restarted the game. The sunk ships were brought back, or "refloated." Van Riper was told, "You are playing out of character." The enemy force was put under rules that let the American plan go ahead, and Van Riper stepped down as its commander, saying the rest of the exercise was scripted.

Joint Forces Command's own report later said: "As the exercise progressed, the OPFOR free-play was eventually constrained to the point where the end state was scripted." OPFOR is the opposing force.

Afterward, the organizers said the exercise was testing new concepts and was "not about winning or losing." Micah Zenko sets out both sides in his 2015 account for War on the Rocks, and part of the organizers' case is fair.

But Van Riper had shown how the plan could be beaten, while it was still a game and nothing real had been lost. Before the restart, the useful question was whether it could happen for real. Instead, the ships were refloated and the plan went ahead.

Not everyone on the American side missed the point. Its commander, Army Lt. Gen. B.B. Bell, said the enemy force had "sunk my damn navy." He later called the exercise "a watershed 'eureka' moment in the application of red teaming," and as a four-star general he went on to set up red teams of his own.

There are two lessons in that. Van Riper won by finding a weakness the other side had not protected. The organizers then set aside what he had shown them about their own plan.

Finding the weakness

In early 2006, ASL was trying to win its first customer in parcel delivery. The market ran on small hubs, and three of us sat down to work out how to beat the competitors already in it. Like Van Riper, we went looking for what the other side had stopped questioning.

What we noticed was that the competitors we were up against ran their small hubs without a supervisor. We knew from our own business how much a supervisor mattered. The trouble was that a small hub does not have enough work to pay for a full-time one.

We tried several ideas, including split shifts and part-time managers. The one we settled on was a supervisor who was also willing to drive. In the morning, they organized the crew and got everyone out on time. They ran deliveries for about half the day. Then they came back and closed out the day as the supervisor, making sure everything was done.

We found a way to make the numbers work. We won that first customer, and that was where our parcel business started. By the time I left ASL, it was more than half of the company.

Not falling into the trap yourself

The business version of refloating the ships is familiar. A customer leaves, and the conclusion is that they were never a good fit. Someone on your team raises a risk in a planning meeting, and the answer, in different words, is that they are playing out of character. Sometimes those explanations are true. They can also end the conversation before anyone asks whether it could happen again.

You cannot hire a Van Riper for every decision. You can ask your own people to play the enemy against your own plan, by telling them you have already lost and asking them how it happened. We did that at ASL on a bid we could not afford to lose.

Our third bid

Around 2017, we were heading into our third bid for the business of a global retailer. We had developed a specialized cross-border consolidation model for them. It solved what they considered their biggest problem at the time, and it was the reason we had the account.

The tender was never designed for our solution. It asked for something different from what we provided, so every time the business came up for bid, we had to explain the model again. By the third bid, many of the people we had worked with had been promoted or moved, and the new people did not fully understand how the model was built to solve their problem.

By any normal standard, our team had put together a great bid. They knew it, and I believed it.

We had been the confident favourite before, and lost. We went into the Christmas holidays as the lead on a bid for a chocolate company. When we came back in January, they had stopped returning our calls, and we lost.

Looking back, we had not stayed close enough to what mattered most to them, or to how we measured up against it. On an account this important, I did not want to learn that lesson again. So before we submitted, I brought our salespeople and operations people together for a pre-mortem.

The pre-mortem comes from Gary Klein, a psychologist who has spent his career studying how experienced people make decisions under pressure. He developed it and described it in a 2007 Harvard Business Review article, "Performing a Project Premortem."

I told the team I truly believed they had done a great job. Then I asked them to imagine that I had just walked in and told them the retailer had called, and we had lost the account. What would your gut tell you? What would you regret?

The meeting ran an hour to an hour and a half. People heard the question as a verdict on their work, and for most of that time my team defended the bid against a loss that had not happened yet. I kept explaining that we were doing this for a specific reason, and that it was not because anything was wrong with the bid.

That surprised me. Getting people willing to share their regrets took most of the meeting. Collecting the regrets took a fraction of that. If I ran one again, I would plan for that from the start. Rushing to the list would have produced polite answers.

The second surprise came once they were past that. The answers flowed easily, and nearly everyone had at least one regret. They sounded like this: I should have called that person one more time. I should have asked them what was really going on. I should have checked in with their technology lead to make sure they understood what we were trying to achieve.

Many of them were about staying close to the customer, the lesson the chocolate bid had taught me.

We listed 14 or 15 regrets on a bid that was already strong. We put them in order of priority, and we dealt with every one of them before the bid went in.

We won.

I will never know whether we won because of that meeting. What I do know is that if we had lost, we would not have been left with 14 or 15 things we wished we had done.

We used it on a sales bid, but it works on any decision you are about to commit to: choosing a new software system for the company, a go or no-go on an acquisition, or launching something new. You do not need a big team either. In a smaller company, four or five people is enough.

Van Riper looked at the American plan the way an enemy would. On the small hubs, we did that to our competitors. Before the retailer bid, we did it to ourselves.

Think of the biggest decision in front of you right now. If you learned a year from now that it had failed, what would you already regret?

Sources

Frequently asked questions

What happened in the Millennium Challenge 2002 war game?

In the summer of 2002, the U.S. military ran a war game called Millennium Challenge that cost about $250 million. Paul Van Riper, a retired Marine Corps lieutenant general, played the enemy. He sent orders by motorcycle courier and launched aircraft using light signals instead of radio, so the American side had very little to see. Then he struck first, and the opening strike sank at least 16 American warships, including an aircraft carrier. The organizers restarted the game, "refloated" the ships, and put the enemy force under rules that let the American plan go ahead.

What is a pre-mortem?

A pre-mortem is a method developed by the psychologist Gary Klein and described in his 2007 Harvard Business Review article, "Performing a Project Premortem." Before a decision is final, the team imagines it has already failed and explains what went wrong. Cole Dolny's version asks the team what they would regret if they had just been told they lost.

How do you run a pre-mortem before a major sales bid?

At ASL, Cole Dolny brought the salespeople and operations people together before the bid went in. He told them he believed they had done a great job, then asked them to imagine the customer had just called to say they had lost the account. What would your gut tell you, and what would you regret? The team listed 14 or 15 regrets on a bid that was already strong, put them in order of priority, and dealt with every one before submitting. ASL won the bid.

Why do teams resist a pre-mortem?

People can hear the question as a verdict on their work. In Cole Dolny's experience, getting the team willing to share their regrets took most of a meeting that ran an hour to an hour and a half, and collecting the regrets took a fraction of that. Plan for that time from the start. Rushing to the list produces polite answers.

How do you find a weakness in a competitor's business model?

Look for what the other side has stopped questioning. In early 2006, as ASL tried to win its first parcel delivery customer, the competitors it was up against ran their small hubs without a supervisor, because a small hub does not have enough work to pay for a full-time one. ASL's answer was a supervisor who also drove: organizing the crew in the morning, running deliveries for about half the day, then closing out the day. ASL won that first customer, and that was where its parcel business started.

Can a pre-mortem be used for decisions other than sales?

Yes. It works on any decision you are about to commit to, such as choosing a new software system for the company, a go or no-go on an acquisition, or launching something new. It does not need a big team. In a smaller company, four or five people is enough.

Cole Dolny is the founder of 6S Advisory Inc. and a Vistage Canada Chair. He works with CEOs and owners on leadership, talent systems, and decision-making, helping them build healthier, more profitable businesses.