A Decision Making Framework You Will Actually Use

You have two offers on the table, or two candidates, or two ways to spend the next year, and you have been going round in circles for a week. What you are looking for is a decision making framework, something that turns the loop into a choice. Most of the ones you will find are too heavy to use on a real Tuesday.

The one worth having fits on a single page and asks five questions in order. Can you undo the decision? How much do you need to know before you make it? What do you want from it? How could it fail? When will you check how it went? Answer those, in writing, and you will have decided more carefully than most organisations do.

A Decision Making Framework Has to Be Short

The frameworks that exist in books tend to have grids, weightings, scoring scales and acronyms. They are built to look rigorous, and they are fine for a committee with a week to spare. For a founder choosing between two suppliers on a Wednesday afternoon, they get read once and never opened again.

A framework earns its place by being used. That means it has to take less than half an hour, run on paper, and work when you are tired. If you have ever made a poor call late in a long day, the piece on decision fatigue explains why, and why the answer is a fixed process rather than more willpower. The five questions below are built to be a fixed process.

Can You Undo It

Start here, because the answer decides how much effort everything else deserves. Some decisions are cheap to reverse and some are not, and treating them alike is the most common way to waste time.

Jeff Bezos put it in his 2016 letter to Amazon shareholders this way: “Many decisions are reversible, two-way doors.” Those, he argued, can use a light-weight process. The full letter is worth reading for how he applies it to a large company.

The other kind are the ones you cannot easily walk back: a long lease, a co-founder, a large purchase, several years of your life spent on a qualification. Those deserve the full five questions. The reversible ones deserve two minutes and a bias toward acting. Sorting them at the start saves you from agonising over the small ones, and from rushing the big ones.

Set the Information Threshold and the Deadline

Ask what you would need to know to be comfortable, and how long it is worth waiting to learn it. Then write down a date.

Bezos made the same point in that letter. Most decisions, he said, should probably be made with somewhere around 70% of the information you wish you had. Wait for 90% and you are usually slow. His reasoning is that if you get it wrong, you can correct course quickly, and that being slow is costlier than being occasionally wrong. That holds for reversible decisions. For the hard-to-undo ones, you should wait longer and gather more, but you still need a date, or the research becomes a way to avoid choosing.

The deadline is the part people skip. Without one, more information always looks like a good reason to wait. With one, you are forced to ask what is still missing and whether you can get it in time. If you cannot, you decide with what you have.

Write the Criteria Before You Look at the Options

This is the step that does the most work, and it takes ten minutes. Before you compare the choices, list what you want from the decision. Three to five items is enough. Rank them.

The order matters. Once you are looking at a specific option, you start to invent reasons it is the right one. You notice the strengths that suit you and explain away the weaknesses. Writing the criteria first fixes the test before you know who is taking it.

Then score each option against your list, roughly, with words if numbers feel false. You are not looking for a total. You are looking for the place where the option you are drawn to loses on something you ranked highly. That is where the real discussion is. The article on structured decision making goes further into why this beats trusting your judgement in the moment.

Run a Premortem

Once you lean one way, try to break it. The most practical method I know of is the premortem, described by the psychologist Gary Klein in the Harvard Business Review in 2007.

You imagine that it is a year from now and the decision has failed badly. Then you write down every reason you can think of for the failure. Klein cites research from 1989 by Mitchell, Russo and Pennington finding that imagining an outcome has already happened increases people’s ability to correctly identify reasons for it by 30%. Saying the plan has already failed gives people permission to name the doubts they would otherwise keep to themselves.

Alone, the version is short. Write the failure story, list the reasons, and mark the two that seem most likely. Then change the plan to address them, or decide that you accept the risk with your eyes open. Either outcome is better than finding out later.

Record It and Set a Review Date

The last step takes five minutes and is often skipped. Write down what you decided, why, what you expected to happen, and the date you will look again.

There are two reasons. The first is that memory rewrites itself. After an outcome arrives, it is very easy to believe you saw it coming, or that you never really wanted the other option. A note made at the time keeps you honest about the quality of your reasoning, which is different from the quality of the result. A good decision can turn out badly, and a bad one can get lucky. Only the record lets you tell them apart.

The second is that a review date gives you permission to stop worrying. You have made the call and you have a time to check it. Until then, the job is to carry it out.

The One-Page Version

Here is the whole decision making framework as you would write it on a single sheet. First line: the decision, in one sentence. Then the five answers, each in a line or two. Can it be undone, and if so how cheaply? What do I still need to know, and by what date do I decide anyway? What do I want from this, ranked? If it failed, why? When do I look again?

That is the entire page. It should take between ten minutes for a small decision and an hour for a large one. If it takes longer, the decision is probably two decisions, and you should split it.

Keep the sheets. After a few months they become the most useful record you have of how you actually think, including the places where you keep making the same kind of mistake.

Practise on a Small One First

Do not save the framework for the biggest choice you face. Run it first on something small and reversible, such as which of two tools to try, or how to organise next quarter. The steps feel slow and mechanical the first time, and you want that to happen when little is at stake.

After three or four small runs, the questions start to arrive on their own. You catch yourself asking whether you can undo something before you have opened a spreadsheet. By the time a large decision comes, the process is already habit, and you spend your attention on the decision instead of the method.

Three Decisions Run Through It

Take a business owner whose books have fallen behind, deciding whether to hire a bookkeeper or an accountant. Can it be undone? Largely yes, since you can change adviser. So the process is light: set a short deadline, write down what you need, and pick. The article on bookkeeper vs accountant gives the criteria most small businesses end up with.

Now take someone weighing an MBA. Can it be undone? Barely. The time is spent and the fees are paid. This one gets the full treatment: criteria written first, a premortem on the version where the degree does not open the door you expected, and a review date. If you are in that position, is an MBA worth it is the place to build the criteria.

A career move sits between the two. Leaving a salaried job to advise companies can be walked back, at a cost in time and reputation, so it earns the criteria and the premortem but not months of research. Our guide on how to become a consultant covers what belongs on that list. And if a choice like this has dragged on long enough to cost you sleep, the advice on how to manage stress at work pairs well with the deadline step. If the decision is really a way out of exhaustion, read about recovery from burnout first, because choices made at the bottom tend to repeat the last one.

Where the Framework Stops

A framework improves the reasoning. It cannot supply the values. If two options score the same, the difference is what you care about, and no page can tell you that.

It also cannot remove the discomfort of choosing. Once you have decided, the useful move is the one Bezos described as disagree and commit: if others on your team lean the other way, hear them, decide, and then back the decision fully. Half-hearted execution can sink a reasonable choice.

So keep the page short and keep it somewhere you will find it. Five questions, answered in writing, before you decide. The next time you catch yourself going round in circles, the loop is a sign that one of the five has not been answered yet, and you can go and find which.