How Leaders Should Make Critical Decisions

Bias for action: That’s what we are told. But we are also told to think before we act. That’s why it is key to spot decisions that have minimal consequences. They are the ones that require less debate. You should spend time only on decisions that lead to big outcomes.

Jeff Bezos of Amazon likes to classify decisions as type 1 or type 2. Type 1 choices are irreversible. They are one-way doors. If you walked through and didn’t like what you saw on the other side, Bezos wrote, you couldn’t get back to where you’d been before. He said that when making such choices, you had to be methodical, careful, and slow, with great deliberation and consultation. You could not move fast and break things. Not even at Amazon.

But Bezos also noticed that most decisions weren’t like that. Many were reversible, like two-way doors. You could back out if you wanted to. You didn’t have to live with the consequences for that long. When making type 2 choices, decide quickly or delegate to a small group of smart individuals. You’ve heard this before: Embrace empowerment, autonomy, and experimentation. Fail fast to succeed early.

The key, therefore, is to be clear on the decision types. Your team must avoid one-size-fits-all thinking. There’s no room for intellectual laziness. When making irreversible decisions, dissent is your friend, not your foe. You don’t want unthoughtful risk aversion. But you can’t afford groupthink either.

M&A: The One Thing You Need to Get Right

Mergers and acquisitions are one-way doors. And companies fare poorly: 70% to 90% of M&As end in failure, and that’s the general conclusion of most studies.

This has many causes. A company doesn’t normally acquire other firms very often. The general lack of experience means that managers only see potential deals brought up by bankers and private equity, whose single motivation is to make deals. But more troubling is how decisions on M&A get made.

Imagine that you work at a manufacturer making control systems for aircraft cockpits. A business leader spots another company that makes control systems for cabins. So they think that an acquisition makes sense. For the first time, the company can offer customers — the airlines — a combined solution: one single control system for the entire aircraft. A brilliant strategy.

In a decentralized organization, the business leader would directly negotiate with the seller. They then would submit a financial proposal for corporate headquarters to approve. But here’s the thing: How do you judge if the deal was overpriced or not?

Typically, there’s a finance team from headquarters that conducts due diligence, investigating whether the company you want to buy is really up to what’s claimed on paper. It wants to make sure there are no hidden legal liabilities and no shady business involved. But the fundamental nature of the negotiation remains unchanged. It’s the person who found the deal to follow up and negotiate with the seller.

This is bad. Because it’s hard for the same person to walk away from the deal. After months of hard work, they have already fallen in love with the company they want to buy. This is classic escalation of commitment. People stick to their guns. It’s the price we’ll pay to execute our strategy, they’ll say.

This makes no sense, of course. If you overpay in an acquisition, your strategy will never see the expected financial benefit. The amortized cost will eat up all the future profit you seek to create in the first place.

The only way to avoid such irrational behaviors is to make the dissenting voices heard. Separate deal-making from deal-negotiating. Split the responsibility.

Honeywell booth at the 3rd China International Import Expo (CIIE) in Shanghai

Promoting the De-Escalation of Commitment

That example of an aircraft control system was real. It was Honeywell’s acquisition of Baker Electronics in 2002. Honeywell overpaid and ended up with a $20 million write-off a few years later. Shortly afterward, then CEO David Cote installed a new system. It was the overhaul that finally turned Honeywell into a world-class M&A machine.

Business leaders would still identify candidates for acquisitions. Cote actually pressed them even harder to aggressively scour the market. He wanted to build a broad pipeline of acquisition targets. “You have to kiss a lot of frogs before you find your prince,” he wrote in his memoir.

Things get interesting once the target was identified. After the potential deal was put together, the business leaders would step back. The corporate M&A team takes over the actual negotiation. The business leaders didn’t like it at first. They felt like they were losing control. But Cote noted it was a lot easier for someone to walk away from a deal when they were not the one who’d put it together hoping that it’d work out.

This idea of promoting alternative viewpoints is obviously not only applicable to M&As. It’s crucial whenever decisions aren’t reversible.

Farming for Dissent at Netflix

At Netflix, employees use a shared spreadsheet when proposing an idea. You invite dozens of colleagues for input. They then rate your idea on a scale of -10 to +10 with explanations and comments. CEO Reed Hasting uses this method to get clarity. The spreadsheet system is simple, and it gathers assent and dissent, he noted. That’s how Hasting collects feedback before making important decisions.

He learned his lesson the hard way. It was the biggest debacle in Netflix’s history. Back in 2007, when DVD rental by mail was still a gigantic business, Netflix would charge $10 a month for DVD mailing and online streaming. Hasting foresaw the importance of the streaming service. He understood Netflix should become ready for the imminent shift in consumer behaviors. He also wanted the streaming operation to focus on innovation, and the DVD rental to focus on efficiency. So he decided to split up the company.

Reed Hasting, CEO and founder of Netflix

Hasting created Qwikster, which charged $8 a month for DVD rental. Netflix would provide online streaming only for $8. Customers who needed both thus paid $16 in total. It’s a price hike.

Customers hated it. Millions of subscribers left. Netflix stock dropped by 75 percent. Months would pass before the company recovered. In the post-mortem, the managers confessed the CEO had been so intense in his belief that they’d felt powerless. They’d been afraid to speak up. “I should have laid down on the tracks screaming that I thought it would fail,” said one VP. “But I didn’t.”

These observations aren’t meant to make us suffer analysis paralysis. But we need to be clear about what types of decisions we are making. Ask yourself:

1. Among the decisions that you are making this coming week, which are two-way doors? For those type 2 decisions, explain to others that you can back out if they don’t work out. Relax the control. Give power to those who actually do the work.
2. For type 1 decisions that are irreversible, did you check with people with alternative viewpoints? Did you actively farm for dissent?

If you don’t want to farm for dissent online, you can do it the old fashion way. Put out the situation you face before your team members, as well as your recommended actions. Let them come to their own conclusions. You divide the team into breakout groups for deliberations to avoid groupthink. Sometimes your team might agree with what you’ve proposed. Other times they generate solution you may not have thought of — and that’s great.

Remember, you are not building consensus here. You still own the consequences of your choices. You are checking your biases. Am I increasing commitment because I have fallen in love with my own ideas?

Like what you have read so far? Check out more on my blog.

P.S., What’s your company’s approach to ensuring high decision quality? If you have tips and methods that help with your own decision-making, please share them with us. We’d love to hear your experiences.