Decision-Making: Methods and Biases

Leaders are paid to make decisions, often with incomplete information and under time pressure. Good decision-making isn't about always being right; it's about using a sound process: matching the speed to the stakes, involving the right people, comparing options openly and guarding against the predictable errors every human brain makes. This lesson gives you a toolkit for each of those steps, with a worked example from a Chicago warehouse facing a peak-season backlog.

Step 1: How reversible is it?

In his 2015 letter to Amazon shareholders, Jeff Bezos split decisions into two types. Type 1 decisions are "one-way doors": consequential and nearly irreversible, so they deserve slow, careful deliberation. Type 2 decisions are "two-way doors": if you're wrong, you can walk back through. He warned that as organizations grow, they tend to use the heavy Type 1 process for everything, which makes them slow.

Step 2: Who decides?

Confusion about decision rights causes more friction than bad decisions do. Bain & Company's RAPID framework assigns five roles, ideally written down before the discussion starts:

This connects directly to the delegation levels in Lesson 6. Say it out loud: "Luis, you'll recommend; I'll decide by Friday; I want input from both shift leads."

Step 3: Compare options with a weighted decision matrix

When several options and several criteria compete, a weighted matrix makes the trade-offs visible. List the criteria, give each a weight that adds up to 100%, score each option from 1 (poor) to 5 (excellent) and multiply:

Weighted score = Σ (criterion weight × option score)
Excel: =SUMPRODUCT($B$2:$E$2,B3:E3)

Worked example: clearing a peak-season backlog

Dana Kowalski's Chicago warehouse is falling behind as the holiday season approaches. Her team scored four options against four criteria:

Option D: 0.35×3 + 0.25×4 + 0.20×3 + 0.20×4
        = 1.05 + 1.00 + 0.60 + 0.80 = 3.45

Two lessons from this example. First, the best option often appears during the scoring: D didn't exist until the team saw that A and B each had a fatal weakness. Second, check how sensitive the result is to the weights. If Dana weighted speed at 40% and cost at 20%, B and D would tie at 3.60. The matrix doesn't make the decision; it shows you exactly what you're trading off. Pick-to-light scored lowest for this season's problem, but its accuracy score makes it worth considering as a separate, longer-term investment.

Step 4: Watch for biases

Psychologists Daniel Kahneman and Amos Tversky spent decades documenting systematic errors in human judgment. In Thinking, Fast and Slow (2011), Kahneman describes two modes of thinking: System 1, fast and intuitive, and System 2, slow and deliberate. Most biases come from letting System 1 answer a question that needs System 2. These six show up constantly in management decisions:

Step 5: Stress-test with a pre-mortem

Psychologist Gary Klein described the pre-mortem in a 2007 Harvard Business Review article. Before committing to a decision, the team imagines it's a year later and the decision has failed spectacularly, then each person writes down why. It makes it safe to voice doubts that people otherwise keep to themselves. A script:

"Imagine it's next February. We went with Option D and it was a disaster.
Take three minutes and write down every reason why it failed.
Then we'll go around the room, one reason each, until we've heard them all."

A simple debiasing checklist for important decisions:

• I know whether this is a one-way or a two-way door.
• Roles are clear: who recommends, gives input and decides.
• We compared at least three real options, including "do nothing."
• People wrote their views before the group discussion.
• We ran a pre-mortem and added safeguards for the top risks.
• We agreed on what evidence would make us revisit the decision.

Announce the decision

A well-announced decision is half implemented. Use this structure: "We decided (decision). We chose it because (top two reasons). We also considered (other options) and didn't choose them because (one line each). We'll revisit it if (trigger). Here's who does what by when. If you disagreed, I heard you, and I'm asking you to commit to making this work."

Summary

• Match speed to reversibility: slow down for one-way doors, move fast on two-way doors (Bezos, 2015).
• Clarify decision roles with RAPID: recommend, agree, perform, input, decide.
• A weighted matrix (Σ weight × score) makes trade-offs visible; always test how sensitive it is to the weights.
• Watch for confirmation bias, anchoring, sunk costs, availability, overconfidence and groupthink.
• Run a pre-mortem, collect independent views first and announce decisions with the reasons and a review trigger.

Next lesson: Leading Remote and Hybrid Teams. You'll set up working agreements, handle time zones and build trust when your team rarely shares an office.