How to make decisions under uncertainty

Here's how to make decisions under uncertainty: stop trying to predict the outcome and start judging the bet. Figure out what you know, what you can't know, and what you could find out cheaply. Then pick the option that holds up reasonably well across several possible futures, protect yourself against the worst one, and build in a way to adjust once you learn more.

You will never get certainty on the choices that matter most: a new job, a move, a relationship, a business. The good news is you don't need it. You need a decision you can defend with the information you had, plus a plan for what you'll do if things go differently.

Why uncertainty feels so paralyzing

Uncertainty isn't the same as risk. With risk, you roughly know the odds, like a weather forecast. With true uncertainty, you don't even know the odds. Will this startup survive? Will you like living in Denver? Will this relationship last? Nobody can give you a percentage.

Our brains don't love that. Many people would rather take a known, mediocre option than an unknown one that might be better. That instinct is sometimes wise and sometimes just fear dressed up as caution.

There's also a trap called outcome bias: judging a decision by how it turned out instead of by how it was made. If you drive home after a drink and nothing happens, it was still a bad decision. If you take a sensible job and the company folds, it was still a reasonable choice. Once you accept that good decisions can have bad outcomes, uncertainty gets a lot less scary. You're responsible for the quality of your thinking, not for controlling the future.

Step 1: Separate what you know from what you don't

Grab a page and make three columns:

• Known: Facts you're confident about. The salary, the rent, the commute, your savings.
• Knowable: Things you don't know yet but could find out. What the team is like, what the neighborhood feels like at night, what your partner really thinks.
• Unknowable: Things nobody can tell you. Whether the company will be acquired, whether you'll meet someone in the new city, whether the market goes up.

This simple sort does two things. It shows you where homework would actually help. And it lets you stop spending energy on the unknowable column, because no amount of worrying will move anything out of it.

Step 2: Buy information cheaply

For every item in the "knowable" column, ask: what's the cheapest, fastest way to learn this?

• Talk to two people who've done the thing you're considering.
• Visit for a weekend before you move.
• Freelance on the side before you quit.
• Rent before you buy.
• Ask the hiring manager directly about the things you're nervous about.

Then ask a second question: would the answer change my decision? If you'd make the same choice either way, skip it. Research that can't change your mind is just procrastination with better branding.

Step 3: Think in scenarios, not predictions

Instead of guessing what will happen, sketch three versions of the future for each option:

• Good case: Things go well, but realistically, not in fantasy.
• Likely case: The most ordinary version of how this plays out.
• Bad case: Things go wrong in a plausible way.

Now look across them. You're hunting for the option that's acceptable in all three, not the one with the most dazzling good case. A choice that's great if everything breaks your way and disastrous otherwise is a fragile bet. A choice that's fine in the bad case and good in the likely case is usually the stronger one.

This is a gentler cousin of a technique called a premortem, where you imagine a decision has already failed and work out why.

Step 4: Check the downside you can't recover from

Some bad outcomes are annoying. Others are ruinous. Uncertainty is far more manageable when nothing on the table can wreck you.

Ask of each option:

• What's the worst realistic thing that could happen?
• Could I recover from it? How long would it take?
• What would I need in place to survive it: savings, a fallback, a person to call?

If an option has a small chance of an outcome you truly couldn't recover from, that deserves extra weight, even if the odds seem low. For decisions involving significant money, health or legal exposure, this is the moment to talk to a qualified professional about your specific situation.

Step 5: Make it easier to change course

Uncertainty is hardest when a decision locks you in. So look for ways to loosen the lock:

• Start smaller. Test a side business before going all in.
• Negotiate an exit. A shorter lease, a trial period, a return policy.
• Stage the commitment. Agree to six months, then reassess.
• Keep a bridge. Leave on good terms so the door stays open.

Turning a one-way door into a two-way door is one of the most powerful moves available. The guide on reversible vs irreversible decisions goes deeper on this.

Step 6: Set a decision date and a review date

Uncertainty never runs out on its own. There will always be one more thing you could learn. So pick a date to decide, and separately, a date to review how it's going.

The decision date stops you from waiting forever. The review date gives you permission to act now, because you know you'll check in later with real information instead of guesses.

A worked example: Rosa and the out-of-state offer

Here's a hypothetical. Rosa, 31, works in marketing in Ohio. She's been offered a role in Portland with a modest raise and a bigger title. She's never lived in the Pacific Northwest, her friends are in Columbus, and she's not sure the new company is stable.

Known: The salary bump, the moving costs the company will partly cover, the rent range in Portland, her current savings.

Knowable: What the team is like, how the company is doing financially, whether she'd like the city.

Unknowable: Whether she'll make close friends, whether the company exists in five years, whether she'll meet a partner there.

Cheap information: She asks for a call with two future teammates and asks them directly about turnover and funding. She spends a long weekend in Portland using vacation days. She asks a former colleague who moved across the country what the first year was like.

Scenarios:

• Good case: She loves the job, builds a social circle in a year, and grows quickly.
• Likely case: The job is solid, the first winter is lonely, and by year two she has a few close friends.
• Bad case: The company has layoffs within a year, and she's job-hunting in a city where she has no network.

Downside check: The bad case is painful but survivable. She has about four months of expenses saved, and her title would help her land something else. Her old manager has said she'd be welcome back.

Loosening the lock: She negotiates a later start date so she can sublet instead of signing a 12-month lease right away.

Dates: She decides by Friday. She sets a review date for nine months after the move.

She takes the job. Not because she knows it will work out, but because she's checked that even the bad case is something she can handle, and she's left herself room to adjust.

Common mistakes under uncertainty

• Waiting for a sign. Certainty isn't coming. Neither is a sign. Set the date.
• Overweighting the vivid. One friend's horror story about a move feels like data. It's one story.
• Ignoring the cost of not deciding. Staying put is also a bet with its own unknowns.
• Collecting information that can't change your choice. If the answer wouldn't move you, skip the question.
• Judging yourself by the result alone. If you made a careful choice and it went badly, that's bad luck, not bad judgment. Our guide on what to do after making the wrong decision helps with that part.

Simple tools for small uncertain choices

Not every uncertain decision deserves six steps. For low-stakes choices with genuinely unknown outcomes, like which of two restaurants will be better tonight, the information simply isn't there. A quick coin flip is a perfectly rational way to settle it, and your reaction to the result often tells you which one you wanted.

For bigger ones, it can help to talk it through with someone who isn't emotionally invested. Choice can give you a clear recommendation based on the trade-offs you describe, and the companion can look at photos you share, like a listing or an offer letter, if that's easier than explaining every detail.

The bottom line

You can't remove uncertainty from big decisions, but you can make it manageable. Sort what you know from what you can't. Learn the cheap things. Picture a few futures instead of betting on one. Make sure the worst case is survivable, keep a way to change course, and set a date. Then decide, knowing that a well-made choice is the only part you ever really control.