A Decision-Making Framework for Leaders: Nine Steps for the Calls That Matter

A decision-making framework for leaders is a fixed order of tests that every important call passes before anyone commits. This one has nine: confirm the decision matters, define the problem, set a checkable outcome, generate at least three options, score them on a rubric, apply your philosophy, red-team the choice, plan it with tripwires, and record it.

Each step has one test, and if the test fails you don’t advance. That rule stops a leadership team from debating options for a problem nobody has defined, and it means the same information produces substantially the same decision no matter who’s in the room.

How Do CEOs Make Decisions That Hold Up?

They run every important call through the same tests in the same order, write down what they decided and why, and grade the process they used as well as the result. That consistency lets a leadership team make substantially the same call the CEO would, given the same information, without the CEO in the room.

The grading matters because outcomes mislead. You can make a crappy decision that still produces results; count your blessings and don’t repeat it. You can make a sound decision that doesn’t work; given the same information and constraints, you’d make it again.

Few companies make decisions well. In McKinsey’s 2019 report on decision making, only 20 percent of respondents said their organizations excel at it. Jerry’s line is that leaders create a predictable environment for people to thrive, and a decision-making framework is that environment for the hardest calls.

The Decision-Making Process Steps, One Test Each

Run the steps in this order. If a step’s test fails, stop and fix it before moving on.

Step	The single test	What fails if you skip it
1. Is it important?	The problem is real, matters to the company, is specific, and needs attention now	You spend leadership time on a story, or take a call that belonged to your team
2. Define the problem	You can state it in two sentences, with evidence and precedent	You fix the symptom and pay for the problem twice
3. Set a clear outcome	The outcome is time-bound, tangible and checkable later	Nobody can say afterward whether it worked
4. Generate options	At least three real options are on the table	You pick between the first two you happened to see
5. Apply the rubric	The chosen option passes all eight rubric checks	A choice that wins this quarter and breaks next year
6. Apply the philosophy	The choice fits the trade-offs the company says it accepts	Every decision reopens the argument about risk and speed
7. Red-team it	Two named skeptics and a pre-mortem can’t sink it	The flaws surface after the money is spent
8. Plan, tripwires, go	An owner, milestones, dates and one to three tripwires are named	The decision stays locked in after the facts change
9. Record it	It’s written down on the day, including your confidence	Hindsight rewrites the story and nobody learns

Step 1: Is the Decision Important?

Before anyone spends a minute on options, agree that the decision deserves its own time. Five checks:

• Real. There’s evidence for the problem beyond someone’s account of it.
• Important to the company. It matters to the business, and you’d still care if nobody saw you solve it.
• Specific. “Sales is a mess” creates anxiety. “We lost the last two renewals to the same competitor” creates something to act on.
• Now. It needs attention now.
• Risk or uncertainty. You know whether the odds can be known. If they can, run the numbers. If they can’t, don’t fake precision; simple rules of thumb often do better. The framing is from Gerd Gigerenzer’s Risk Savvy.

If any answer is unclear, the team is still processing emotion: sharpen the question or table it.

Jerry’s quick filter is the intersection test. Traffic collides where roads cross, and decisions do too. If a decision touches more than one department, it almost always rises to the leadership team. If it lives inside one department, it belongs to the person who runs that department, as long as they have the authority to match. When someone is responsible for an outcome but can’t decide, that gap is the Decision Delta, and closing it is your job.

The ego check is the one leaders skip. Jerry calls the failure Swooping In Syndrome: the leader swoops in, solves the problem, disempowers the team and sets up a co-dependent relationship. It feels good in the moment, and the next time that problem shows up, it comes straight back to your desk.

If the decision can be walked back cheaply, it may not need all nine steps. The Two-Door Decision Test sorts that in a minute or two.

Step 2: Define the Problem

Be rational here. State the problem as a claim, then make it earn its place: what you believe the problem is, why it’s true (facts and figures), when it’s happened before (precedent), and why it matters to the company.

Jerry learned this step the expensive way. Early in his time as a leader, he believed confidence came from other people’s approval. When the customer support team got buried in complaints about slow responses, he asked the other department heads what they’d do. They suggested adding staff during peak hours, the support team agreed, and it worked: wait times fell and customer feedback improved. He figured he had this leader thing locked down. Then staffing went back to normal, the complaints climbed again, and the support team’s morale dropped. The real problem was an outdated call management system that couldn’t handle the volume. The fix had masked the symptom, added cost and left the cause untouched.

Or say turnover is up. The claim might be that a lack of growth opportunities is driving people out. The evidence lives in exit interviews, engagement surveys and the pattern of who left. The precedent is whatever happened the last time the company changed how it develops people. The importance is recruiting cost plus the work that stalls while seats sit empty.

The test: you can write the problem in two sentences. If you can’t, the team hasn’t defined it yet. That’s the trap the customer support fix fell into: the answer arrived before anyone had defined the problem.

Martin G. Moore, author of No Bullsh!t Leadership, puts root cause first in his approach to making decisions quickly: treat the symptom right away if you must, then plan the fix for the cause, or the problem comes back.

Step 3: Set a Clear Outcome

Convert the problem statement into an outcome that is time-bound, tangible and checkable later. A weak outcome: “we want lower turnover.” A strong one: “by the end of next year, regretted resignations on the sales team are half of this year’s, and growth stops being the top reason in exit interviews.”

Write down how the solution has to work inside what already exists, too. A new career development program has to survive contact with your compensation structure.

Step 4: Generate Options

There are always more roads than the first two you see. Avoid the pro/con trap: two options and a list of pros and cons flatten the choice before it starts. Require at least three; five is better. For turnover, the range might run from a better benefits package to more flexible working conditions to a real career development program.

Check your history, because what worked last time pulls hard and the business may have changed. Bring in the people closest to the work, because blind spots are the reason to involve the team. And give it a night: good options often show up on a dog walk or just before sleep, once the arguing stops. If the room lands on two options fast, that’s the signal to look for the third.

Step 5: Apply the Decision Rubric

The rubric turns a list of options into a choice you can defend. An option passes a check when the statement is true:

1. Vision, values and objectives. It fits what you said you’re building.
2. Long term. It’s the clearly better option over years: financially, for the culture, for the infrastructure.
3. Short term. It’s the clearly better option this quarter, on the same three lenses.
4. Reversibility. If it’s wrong, you can reverse it at a cost you can live with, or it has had one-way-door rigor.
5. More time or information. Waiting or learning more wouldn’t change the choice.
6. Worth finding out. If it would, you’ve weighed the cost of finding out against the cost of deciding now, and any research worth doing has a date.
7. 10-10-10. You’d still back it at 10 minutes, 10 months and 10 years.
8. The why. Everyone affected can say why you chose it.

An option that passes all eight is ready to commit. If it fails on five or six, name the specific information you need and the specific date you’ll have it. Waiting with no date is Stealth Stalling. The 10-10-10 check is Suzy Welch’s; Chip and Dan Heath use it in Decisive. The rubric itself comes from Clint Rusch, who now owns Cabinetry & Millwork Concepts, and who talked through making the right decisions as fast as possible with Jerry on episode 44 of the Best Places to Lead podcast.

Step 6: Apply Your Decision-Making Philosophy

Scaling and speed are possible when the rules of the game are clear. Write the philosophy once, when nobody is under pressure, so the hard calls later don’t depend on anyone’s mood that day. It names the trade-offs the company accepts; without it, every decision becomes a fresh argument about the trade-offs themselves. Here’s one, adapted from a real-estate development company’s written philosophy:

Every decision is made through the lens of what creates the best long-term return or mitigates the most risk. We make fast decisions, then iterate if we aren’t getting the result. We don’t mind getting it wrong on the path to getting it right; action beats inaction because it gives us feedback. We believe in failing forward fast. Decisions that affect the structure of the business (vision, values, culture, infrastructure) get more consideration. A $20,000 decision on a single development project is not the same as a $20,000 decision on which CRM we deploy.

Yours will read differently. It has to answer four things: the lens you judge decisions through (return, risk, customer outcome, brand), how you feel about being wrong on the way to being right, how structural decisions get more weight than tactical ones, and the dollar threshold at which a decision gets escalated.

Step 7: Red-Team the Chosen Decision

Before the decision ships, pick your two most skeptical people as designated skeptics: people who’ll challenge the status quo and think like an adversary. Their job is to shoot holes in the chosen option and find every point of failure. They aren’t the decision owner, and they have permission to be wrong and to make the room uncomfortable. On a new product line, they’d test whether the market research is stale, how competitors will respond and whether production can scale.

Then run a pre-mortem, the technique developed by research psychologist Gary Klein. Imagine it’s a year and a half from now and the decision has failed. Walk back from the failure: what happened, and what did we miss? Treating failure as a fact frees people to name causes they’d never volunteer about a plan that only might fail. Each cause goes into the plan as a risk with an owner.

Red-teaming only works on a team that can disagree in the room. If yours can’t yet, build Positive Constructive Conflict first.

Step 8: Build the Project Plan, Set Tripwires, and Go

Roll the decision into a project plan with an owner, milestones, dates, dependencies and risks, all named. Tell everyone affected what was decided and why.

Then add one to three tripwires, Chip and Dan Heath’s term in Decisive for signals you commit to in advance that force the decision back open. “If the pipeline falls below the number we set today, by the date we set today, we reopen the hiring plan.” “If we’re six weeks in and the first milestone is missed, we revisit scope.” Without tripwires, decisions stay locked in after the conditions that justified them have changed. The weekly leadership meeting is where they get checked; the guide to an operating rhythm for a leadership team shows where that meeting sits.

Step 9: Record the Decision

Write it down on the day you decide, because a record written later gets rewritten by the outcome. The field that matters most is your confidence level, captured at the time. Annie Duke calls judging a decision by how it turned out “resulting,” and Thinking in Bets is the case against it. A confidence level written on decision day is how you tell, months later, a good decision with a bad outcome from a bad decision that got lucky.

How Do You Make Big Business Decisions Without Slowing Down the Small Ones?

Run all nine steps on one-way doors and on anything worth half an hour of the leadership team’s time. Smaller, reversible calls collapse to four: is it important (step 1), what outcome do we want (step 3), the rubric (step 5) and the record (step 9). Options, philosophy, red team and plan scale with the decision’s size.

If the problem is speed, read How to Make Decisions Faster as a CEO: the fast no, the open-decision audit and what a slow maybe costs.

What Should We Write Down After a Big Decision?

Write down the problem, the expected outcome, who owned the decision and who supported it, the assumptions, what influenced it, who needs to know, the options you rejected, the risks and safeguards, the timeline and your confidence that day. Later, add whether it was a good or bad decision with a good or bad outcome, and what you’d change.

Try This: take one decision your leadership team will make in the next two weeks and fill in this record on the day you make it. Put the review date on the calendar now, and fill in the last two rows when it arrives.

#	Field	What to write
1	The problem	The two-sentence statement from step 2
2	Expected outcome	The checkable outcome from step 3
3	Owner and supporters	One owner by name; who backed the call
4	Assumptions	What has to stay true for this to work
5	Influences	The data, people or frameworks that shaped it
6	Who needs to know	Everyone who changes what they do because of it
7	Options considered	What you rejected, and why
8	Risks and safeguards	What could go wrong, and what you put in place
9	Timeline	Milestones, tripwires and the review date
10	Confidence today	High, medium or low
11	Decision vs outcome	At review: good or bad decision, good or bad outcome
12	What we’d do differently	Before the decision, in it, and in execution

Use This With Your Team: A 20-Minute Decision Drill

Pick one live decision for your next leadership meeting, ideally a topic that has survived more than one meeting without closing. If your team has a few of those, We Keep Having the Same Conversation and Nothing Gets Decided explains why they circle. Then run the first four steps on the clock:

1. Three minutes: is it important? Run the five checks and the intersection test out loud. If it fails, table it and pick another.
2. Five minutes: define the problem. Everyone writes the problem in two sentences, silently, then reads them aloud. The differences are the real disagreement.
3. Four minutes: set the outcome. Agree on one time-bound, checkable sentence and write it where everyone can see it.
4. Five minutes: generate options. Get at least three on the board before anyone argues for one.
5. Three minutes: hand it off. Name the decision owner and two skeptics, and set the date the owner brings back the rubric, the plan and the record.

You leave with an owner, two skeptics and a date.

Decision rules stick when they’re installed where the plan gets made. In Strategic Planning, two days onsite with your leadership team for $12,500 plus travel, every objective answers the questions this framework asks (what problem are we solving, what do we expect to happen and by when, what are the alternatives, what has to be true) and leaves the room with one owner and a guardrail. If you’d rather have a second set of eyes on the live calls in front of you, Executive Coaching is weekly one-hour sessions, $3,000 a month, month to month. Planning isn’t for a company with no leadership team to put in the room yet; that’s coaching. And neither is for a CEO who wants someone else to make the call. The first step is a 30-minute conversation.