Which Decision-Making Framework Fits Your Decision?

One week, four kinds of decision

In January 1993, a gunman murdered seven people in a fast-food restaurant in Palatine, a suburb of Chicago. Walter Gasior, the police department's deputy chief and its spokesperson, suddenly had several jobs at once: grieving families, a frightened town, a department that still had to run, and reporters and film crews filling the streets. “There would literally be four people coming at me with logistics and media issues all at once,” he recalled ¹.

David Snowden and Mary Boone, who told his story in the Harvard Business Review, saw four different kinds of decision in that one crisis, each needing a different response ¹. Stemming the first panic meant acting at once through the media. Keeping the department running meant following established procedure. Some questions needed experts called in. And the weeks after, when parents were afraid to send their children to school, needed something else again. Gasior set up a forum where business owners, high school students, teachers and parents could share their fears and hear the facts, and let the answers come from the town itself ¹.

The authors spelled out what the wrong choice would have looked like. Treat the town's fear as a simple problem, and he might have said “Carry on,” which would have reassured no one. Treat it as a problem for experts, and he might have called in specialists to say the town was safe, risking a loss of credibility and trust ¹. Both are respectable responses. Each is a familiar way of deciding, applied to the wrong kind of decision.

Before you choose how to decide, decide what kind of decision you have.

That is the idea of this page. Most teams have one way of deciding and use it on everything: the analytical team analyzes, the fast team acts, the consensus team convenes. The picker below asks up to four questions about the decision on your desk, names the framework that fits it, and warns you about the trap that framework sets.

Why teams pick a framework out of habit

Snowden and Boone put the problem in one sentence: managers “rely on common leadership approaches that work well in one set of circumstances but fall short in others” ¹. Their own framework, Cynefin, sorts situations by how cause and effect relate, from the simple, where the answer is known, to the chaotic, where searching for the right answer is pointless and the first job is to act ¹.

Jeff Bezos made the same point about speed in his 2015 letter to Amazon shareholders. Some decisions are one-way doors that deserve slow, careful deliberation. Most are two-way doors, and those “can and should be made quickly by high judgment individuals or small groups” ². As companies grow, he wrote, they tend to use the heavy process on most decisions, and “the end result of this is slowness, unthoughtful risk aversion, failure to experiment sufficiently, and consequently diminished invention” ². He filed it under one-size-fits-all thinking ².

So the habit costs twice. A team that brings its heavy process to every choice gets slow. A team that brings its light process to a choice it cannot undo gets hurt, and Bezos's footnote names the opposite error too: companies that habitually make one-way decisions the fast, light way rarely survive to become large ².

Every framework answers one kind of question. A pre-mortem answers “what are we not saying?” Real options answer “how much do we have to commit now?” A decision rights chart answers “who gets to choose?” Pick one by habit and it will still produce an answer, confidently, to a question nobody was asking.

The four questions, and why they come in this order

Each question rules out frameworks that would be wasted effort before the next one is asked. The routes are Stratrix's judgment, built from what each framework's originators say it is for. They are not a scored test from research, and two careful people can answer the same decision differently.

1. Is the problem who decides?If nobody can say who makes the call, no analysis will settle it. Paul Rogers and Marcia Blenko of Bain found that decisions stall where accountability is ambiguous, and proposed naming each role before the debate ¹⁰.
2. Are events moving faster than your meetings?A crisis needs a loop of acting and looking, not a study. John Boyd's OODA loop (observe, orient, decide, act) was built for a contest against an opponent who moves while you think ⁹.
3. How hard is it to undo?This is Bezos's test ². A choice you could reverse in weeks deserves an owner and a deadline, not a framework.
4. Can anyone say what each option will lead to?If analysis or experts can, the risk is in the plan, and a pre-mortem or a set of gates will find it. If the answer shows up only once you act, the risk is in the future, and the job is to buy information cheaply.

If the room cannot agree on the last answer, the disagreement is the finding. Snowden and Boone have a name for not knowing which kind of situation you are in, disorder, and the picker sends you to Cynefin to settle that before anything else ¹.

The options, side by side

Eight places the picker can send you, each with the situation it was built for and the way it goes wrong when used out of habit.

Two of these are often confused. Stage-Gate and real options both release money in rounds. Stage-Gate, from Robert Cooper's work on new products, is a process for a pipeline of projects ⁶. Real options, in Timothy Luehrman's phrase, treat a strategy as a series of options rather than a single projected cash flow ⁴. Use the first to run many projects the same way, and the second to shape one large bet. Scenario planning comes from Pierre Wack's work at Royal Dutch Shell, and its test was never the elegance of the stories but whether they changed what managers decided ⁸.

Using the picker in the meeting

The picker works best before the meeting where the decision is argued, not during it. Five minutes of it can save an hour of people talking past each other, because most of those arguments are about the kind of decision, not the decision.

1. Write the decision as one questionA topic (“Europe”) cannot be sorted. A decision (“Do we close the German office by January?”) can.
2. Have each person run the picker aloneFive minutes, no discussion. Collect the framework each person reached and the answers that led there.
3. Compare the paths, not just the answersIf one person says the choice is easy to undo and another says it is not, that is the conversation to have first.
4. Run the framework, then the decisionUse the framework the room agrees on, then run the decision itself: one question, real options, agreed criteria and one decider, as in getting a room to decide.

Gasior did not have a picker. He had the judgment to see that one week held four kinds of problem, and to answer each on its own terms ¹. Most teams never get a week that forces the lesson on them, which is the danger: with no crisis to jolt them, they keep using the framework they used last time. Asking what kind of decision this is takes five minutes. Getting it wrong can cost a year.

Questions

What is a decision-making framework?

A repeatable way to reach a choice: a set of questions, steps or roles built for one kind of decision. Pre-mortems, real options, Stage-Gate, scenario planning, Cynefin, the OODA loop and decision rights charts are all decision-making frameworks, each made for a different kind of problem.

Which decision-making framework is best?

None is best in general. The right one depends on whether the real problem is who decides, how fast events are moving, how hard the choice is to undo and whether anyone can predict its results. The picker on this page sorts those four questions.

How do I run the decision meeting once I know the framework?

The meeting still needs one clear question, options that differ in kind, criteria agreed before scoring and one named decider. The Stratrix Craft page on getting a room to decide walks through it step by step.

Is the picker a scientific test?

No. The routes are Stratrix's judgment, drawn from what each framework's originators say it is for. Use the answer to start the conversation about what kind of decision you face, and expect careful people to disagree at first.

Where to go next

• Getting a Room to Decide: A Decision-Making Framework
• Decision Rights Chart Builder
• RACI vs. RAPID vs. DACI