Sort decisions by how reversible they are and decide how much time each one deserves.
Before you read further, list the decisions you face in the next two weeks and note beside each roughly how much time you have spent on it so far. Most people find their time has gone to the loudest decisions, the ones people keep asking about, rather than the ones with the most at stake. This module is about matching your effort to the decision, and it starts with one sorting question: can you undo it?
Hafiz manages operations for a small chain of cafes across Singapore (an invented example, as are the details that follow). On one Monday he has four things to decide: a new seasonal drink, which supplier to use for paper cups, whether to try a new rostering app, and whether to sign a five-year lease for a fourth outlet in Tampines. The chain currently has three.
He spends most of the week agonising over the drink and the cups, because those are what people keep asking him about. The lease is signed on Friday afternoon after a twenty-minute call.
The idea that helps here comes from Jeff Bezos's letters to Amazon shareholders, most clearly the letter covering 2015.
Bezos described one-way doors as decisions that are consequential and irreversible, or nearly so. Once you are through, you cannot come back, so they should be made slowly and carefully, with plenty of discussion. Two-way doors are decisions that can be changed or reversed. If you do not like the result, you can step back out. Most decisions are two-way doors, and they should be made quickly, by individuals or small groups with good judgement, because a mistake costs little and can be fixed.
His warning was that as companies grow, they tend to apply the heavy one-way-door process to every decision, including two-way doors. The result is slowness, timid choices and a lack of experiments. Two-way door decisions should be short and sensible, made by the person closest to the problem. Deciding them quickly means the team learns faster and the decision-maker's attention is saved for decisions that need it.
Almost anything can go wrong, so asking whether something could go wrong will not sort your decisions. Ask instead: if it goes wrong, how much does it cost to reverse, and how long does it take? If reversing costs a little money and a few weeks, it is a two-way door.
In Hafiz's week, if customers dislike the seasonal drink, it comes off the menu in a month, at the cost of some wasted syrup and a few hours of staff training. The cup supplier is a two-way door if the contract allows him to switch with reasonable notice. The rostering app has a free trial. All three are two-way doors, and he could have dealt with them quickly. He could give the drink to his head barista with a budget and a deadline, and pick the cup supplier with the best price and an exit clause, with a date set to review it.
The lease is a one-way door. Once signed, he owes five years of rent whether the outlet does well or badly, getting out early is likely to be expensive, and a wrong call could drag down the whole business. One-way doors are where the tools from the rest of the course belong. Hafiz should have applied the checks from Module 3 to the landlord's footfall figures, and the causal questions from Module 4 to the claim that a nearby MRT exit will bring customers. He should have taken the outside view from lesson 5.3, "Vivid stories and the planning fallacy", to his estimate of how long the outlet would take to break even, and run a pre-mortem, covered in lesson 6.2, before signing.
His real error was the balance of his week: he spent so long on the drink that he had nothing left for the lease.
Other common one-way doors include buying property, signing a long contract, firing someone, making a public commitment that would be damaging to withdraw, choosing a degree that takes years, and anything involving safety or legal exposure.
The opposite mistake, treating everything as a one-way door, is common too, especially in cautious organisations. Every decision goes through the same approval chain. A font change on a slide needs three sign-offs, and a small trial of a new tool waits two months for a committee (both invented illustrations). People stop suggesting changes because getting a yes takes longer than the change is worth. Senior people's attention goes to decisions that do not need them, and the whole team slows down.
Many decisions that feel permanent can be made reversible, or more reversible, with some thought. Before you treat a decision as a one-way door, ask whether there is a smaller first step you could take back. That turns one big irreversible decision into a smaller reversible one, followed by a big one made with better information.
Hafiz could have asked for a shorter first lease with an option to renew, or run a pop-up or weekend kiosk in Tampines for two months before committing. A company can pilot a new system with one team. Someone considering a career change can take a short course or a part-time project in the new field before resigning.
Pilots and trials cost money and time as well. For a decision that is expensive to undo, that cost often buys the most useful thing available: real information before the door closes.
For each decision, then, ask what reversing it would cost and how long it would take. If the answer is a little money and a few weeks, decide quickly, hand it to the person closest to the problem, and build in an exit such as an exit clause, a review date, or a budget and deadline. If it is expensive or permanent, look first for a smaller step you could take back, then slow down, discuss it, and use the checking tools from the rest of the course.
Your own month ahead is full of doors of both kinds, and some of the quiet ones may matter more than the loud ones. The activity below asks you to sort them before they arrive.
List ten decisions you face this month. Mark each one as a one-way or two-way door and note how long you will spend on it.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).