How to Make Decisions When You Don't Have Enough Information

Most of the decisions that shape a life — who to hire, what to build, which opportunity to pursue, when to stop — are made under conditions of genuine uncertainty. You don't have all the information you'd like. You can't predict the outcome with confidence. And waiting for more certainty is itself a choice with its own consequences.

The interesting question is not how to eliminate uncertainty before deciding. It is how to decide well despite it.

The Information You Have vs. The Information You Need

Most poor decisions under uncertainty are not caused by too little information. They are caused by confusion about which information actually matters.

The question "do I have enough information to decide?" is almost always the wrong question. The right questions are: which of my current beliefs about this situation are load-bearing — are the ones that most directly determine what I should do — and how confident am I in each of them?

When you map out the load-bearing beliefs, two things tend to happen. Some of them turn out to be things you actually know quite well, or could find out quickly if you tried. Others turn out to be things that cannot be known before deciding — the market will respond how it responds, the candidate will perform how they perform, the relationship will go how it goes. These are genuinely uncertain and no amount of additional information gathering will resolve them.

The useful action is: gather more information on the first category (the beliefs that are important and that you can actually test), accept uncertainty on the second category, and make the decision based on what you can know.

Reversibility as a Decision Variable

The most useful distinction in decisions under uncertainty is between reversible and irreversible.

For reversible decisions — where you can adjust, undo, or learn from the outcome and try something different — the optimal approach is usually to decide sooner with less information. The expected cost of being wrong is low (you can correct it), and the expected cost of delay is real (time passes, the opportunity changes, inaction has costs of its own). Decide fast, observe outcomes, correct.

For irreversible decisions — or near-irreversible ones, where the cost of undoing the choice is very high — the optimal approach is to gather more information before committing. The asymmetry shifts: being wrong is expensive and uncorrectable, so the value of additional information before deciding is higher.

Most people apply their decision-making resources backwards: they spend significant time on reversible decisions and move too quickly on irreversible ones. The reversible ones feel more manageable because they can be undone, which paradoxically invites more deliberation. The irreversible ones create anxiety, which can push toward premature resolution to stop the discomfort of uncertainty.

In executive search, the candidate assessment is substantially irreversible on the client's side — a poor hire costs significantly to undo. I spend most of my deliberative effort on exactly those judgments. The logistics decisions around search management are largely reversible, and I move faster on them.

The Trap of Apparent Certainty

Under stress and time pressure, the mind tends to construct false certainty. You don't actually know how the market will respond — but under pressure, it feels like you do. You don't actually know if the candidate will succeed in the role — but the interview went well and the certainty feels real.

This apparent certainty is not the same as actual confidence. Actual confidence is calibrated — it tracks how much evidence supports the belief and how much uncertainty remains. Apparent certainty is a psychological state that reduces discomfort by making the decision feel simpler than it is.

One of the most useful habits I have developed is asking, before finalizing a significant decision: what would have to be true for me to be wrong about this? Not to introduce artificial doubt, but to test whether my current confidence level is actually calibrated to the evidence. If I can't articulate a reasonable scenario in which I'm wrong, that's a signal that I've stopped thinking carefully rather than a signal that the decision is genuinely clear.

Decision Quality vs. Outcome Quality

One of the most important concepts in decision-making is the distinction between a good decision and a good outcome.

A good decision is one that was well-reasoned given the information available at the time of the decision. A good outcome is one where things turned out well. These are correlated but not identical.

Good decisions can have bad outcomes — because the world is genuinely uncertain and the information available before the decision was insufficient to rule out the bad scenario. Bad decisions can have good outcomes — because the world is genuinely uncertain and the bad decision happened to work out despite its flaws.

The practical implication is significant: you should evaluate your decisions on the quality of the reasoning that went into them, not just on the quality of their outcomes. If you made a well-reasoned decision that had a bad outcome, the lesson is not necessarily that your reasoning was wrong — it may be that you were unlucky in a decision that was correctly calibrated as uncertain. If you made a poorly-reasoned decision that had a good outcome, you should not become more confident in the reasoning — you were lucky in a decision where the outcome could have gone the other way.

This is harder to maintain in practice than it sounds. Outcome bias is strong: we tend to evaluate decisions based on what happened, not based on the quality of the process. But outcome-based evaluation leads to exactly the wrong learning — penalizing good reasoning that got unlucky and rewarding bad reasoning that happened to work out.

The Personal Framework

For decisions that matter, I use something close to this process:

Identify the load-bearing beliefs — the assumptions that, if wrong, would significantly change the decision. Write them down specifically.

Test the ones that can be tested. Make the calls, gather the data, have the conversations that would either confirm or challenge the most important beliefs.

Accept uncertainty on the rest. Make the decision based on the best current understanding, with explicit acknowledgment of what remains uncertain.

Build in a review point. For significant decisions, plan a specific future moment to assess whether the decision is producing the expected result and whether new information has arrived that would change the assessment.

This is not a way to make perfect decisions. There are no perfect decisions under genuine uncertainty. It is a way to make better decisions, more consistently, and to learn more from both the good and bad outcomes.

Manas Majhi is the founder of Majhi Group and Majhi OS. He makes consequential decisions under uncertainty for a living and thinks about the craft of it.