Decision-Making Under Uncertainty: How to Make High-Stakes Calls With Incomplete Information

In his 2015 letter to Amazon shareholders, Jeff Bezos described two kinds of decisions. Type 1 decisions are one-way doors: you walk through, the door closes behind you, and you cannot come back. Type 2 decisions are two-way doors: if you don't like what you find on the other side, you walk back out. His warning was not that leaders make bad calls — it is that, as organizations grow, they start treating *every* decision like a one-way door. They apply the heavy, slow, consensus-driven process meant for the irreversible choice to the reversible one. The result is an organization that is careful, exhausted, and slow.

I have run businesses across wine, real estate, hospitality, water and distribution for long enough to believe the opposite mistake is just as common: treating a one-way door like a two-way door — signing the irreversible deal on a gut feeling because the meeting was running late and everyone wanted to move on. **Most decision-making failures are not about the quality of the analysis. They are about misclassifying the decision in the first place.**

## Stop waiting for certainty — you will never get it

The instinct under pressure is to gather more information. More data, another analysis, one more opinion. Sometimes that is wisdom. More often it is fear wearing the costume of diligence.

Bezos put a number on it that I have found durable: **most decisions should be made with around 70% of the information you wish you had.** Wait for 90% and, in most cases, you are simply slow. The cost of being slow is invisible — it never shows up on an invoice — which is exactly why it is so dangerous. The deal moves to someone else. The market shifts. The team loses faith that anything will ever be decided.

There is a second, subtler trap. Research on leadership and uncertainty has repeatedly found that leaders systematically underappreciate how random and uncertain the world actually is — we narrate the past as if it were inevitable and then expect the future to be equally legible. It is not. A founder who accepts that the world is genuinely uncertain makes better decisions than one who keeps searching for a certainty that was never on offer.

## The first question: is this door one-way or two-way?

Before anything else, I sort the decision into reversible or irreversible. This single classification changes everything that follows.

- **Reversible (two-way door):** a new label design, a trial shipment to a new market, a pricing experiment in one region, a pilot with a new supplier. If it fails, you absorb a small, bounded cost and walk back out. These should be made fast, by the person closest to the work, with no committee.

- **Irreversible (one-way door):** selling a business, taking on structural debt, entering a market that requires years of fixed commitment, hiring or firing a senior leader, putting the family name on something. These deserve slowness, dissent, and a hard look at the downside.

The discipline is to be honest about which is which — and then to *match the process to the type*. Speed on the reversible, patience on the irreversible. Most organizations get this exactly backwards, and pay for it in both directions: paralyzed on small bets, reckless on large ones.


## Protect the downside before you chase the upside

Optimists start a business; survivors keep one. The reconciliation between those two is a single habit: **before any irreversible decision, I make myself answer one question — if this goes wrong, does it wound us or does it kill us?**

A decision that can only wound you is one you can take on incomplete information, because the cost of being wrong is survivable and the lesson is cheap. A decision that can kill you — that puts the balance sheet, the reputation, or the core business at existential risk — earns a different standard entirely. For those, I want to know not the expected outcome but the *worst plausible* one, and whether we live through it. This is why we keep cash sacred and refuse leverage that only works if everything goes right. The asymmetry matters more than the average.

A practical tool here is the **pre-mortem**, developed by the psychologist Gary Klein. Before committing, you gather the team and imagine it is a year from now and the decision has failed badly. Then everyone writes down *why*. It sounds like a small thing. It is the single most reliable way I know to surface the risk that the enthusiasm in the room was busy hiding.

## Decide, then learn fast

The final discipline is the one that separates decisive leaders from anxious ones: once you have classified the decision, protected the downside, and reached roughly 70%, you decide — and then you treat the decision as the *beginning* of learning, not the end of it.

A reversible decision made quickly and reviewed honestly will, over a year, beat a "perfect" decision made too late almost every time, because you get more cycles of learning. Set the decision, set a date to look at the results, and hold yourself to changing your mind when the evidence says you were wrong. Stubbornness and conviction look identical on the day you decide. They look very different a year later. This connects directly to the work of [long-term thinking](/en/news/why-long-term-thinking-wins-in-business): you can afford to be patient about outcomes precisely because you were disciplined about which doors you walked through.

## Key Takeaways

- Classify every decision first: a reversible two-way door (act fast, push it down to the person closest to the work) or an irreversible one-way door (slow down, invite dissent, study the downside).

- Most decision failures are misclassification — being reckless with one-way doors and paralyzed with two-way doors. Match the *process* to the *type*.

- Decide at ~70% of the information you wish you had. Waiting for 90% usually just means being slow, and the cost of slowness never shows up on an invoice.

- Accept that the world is genuinely uncertain; leaders who stop hunting for unavailable certainty make better calls than those who keep searching.

- Before any irreversible decision, ask whether being wrong would *wound* you or *kill* you — and hold a different standard for the second.

- Run a pre-mortem: imagine the decision has failed a year from now and ask why, before you commit. It surfaces the risk enthusiasm hides.

- A decision is the start of learning, not the end. Set a review date and change your mind when the evidence demands it — conviction and stubbornness look identical until then.

## Frequently Asked Questions

### How do you make decisions with incomplete information?

You will almost never have complete information, so the goal is not to eliminate uncertainty but to decide intelligently despite it. Classify the decision as reversible or irreversible, gather roughly 70% of the information you'd ideally want, protect against the worst plausible outcome, and decide. Then review the result on a set date and adjust.

### What are the methods of decision-making under uncertainty?

The most practical methods are: reversibility analysis (one-way vs. two-way doors), downside-first thinking (would being wrong wound or kill you?), the pre-mortem (imagining failure before you commit), and bounded experiments (small reversible bets that buy information cheaply). The common thread is managing risk and learning speed rather than chasing certainty.

### What are the 3 C's of decision-making?

A widely used framing is Clarify, Consider, and Confirm: clarify exactly what you are deciding and why, consider the real options and their downside, and confirm by committing to a choice and a date to review it. The value is in forcing a stop between "considering" and "committing" so the decision is deliberate, not drifting.

### What is an example of a reversible versus an irreversible decision?

A reversible decision is a pricing test in one region or a trial shipment to a new market — if it fails, you absorb a small cost and walk it back. An irreversible decision is selling a business, taking on structural debt, or putting the family name on a new venture — once done, you live with it. Speed is appropriate for the first; patience and dissent for the second.

### How do leaders deal with uncertainty under high stakes?

Strong leaders separate the size of the stakes from the reversibility of the decision, and they decide rather than wait — but they decide *carefully* when the choice is both high-stakes and irreversible. They protect the downside, invite disagreement before committing, and openly acknowledge what they don't know, which research shows can actually increase how competent and trustworthy a leader appears.

## Make the call

If a decision keeps you up at night, the first question is not "what should I do?" It is "is this a one-way door or a two-way door?" Answer that honestly and most of the anxiety resolves itself: the two-way doors stop deserving your sleep, and the one-way doors get the slow, careful, downside-first attention they were always owed.

To see how this discipline plays out across a group built over more than a century — and why we were willing to walk away from deals that didn't clear the bar — explore [the Manzanos Enterprises story](/en/about) or read why [the best deals are sometimes the ones you don't do](/en/news/best-deal-i-walked-away-from-due-diligence-discipline-of-saying-no).

*Meta description: How leaders make high-stakes decisions with incomplete information: classify reversible vs. irreversible doors, decide at 70%, protect the downside, and run a pre-mortem.*

*SEO keywords: decision-making under uncertainty, decisions with incomplete information, reversible vs irreversible decisions, one-way two-way door decisions, high-stakes decision making, pre-mortem, 70 percent rule, leadership under uncertainty*