The Decision Framework Kit
Close or test every decision open more than a week, aim each big one at the constraint that matters, and find out in a quarter whether your judgment is getting better. One filter, 7 steps, and a journal written before the outcome.
What’s Inside
- Step 0, the Constraint filter (which open decision deserves your time, and which go on the stop-doing list)
- The R.E.V.E.R.S.E. Framework (7 steps: Reversible, Expected value, Verify, Evidence, Risk, Second-order, Execute)
- The 30-minute decision process (the rail, minute by minute, with your own skip line)
- The money math (cost of delay, a survivable-loss line for sizing any bet, and a calibration score for your gut)
- A People Decisions chapter (2 questions, wrong seat before wrong person, and why letting someone go is never reversible)
- The Decision Journal and the decision document (written before the outcome, re-read before you re-decide)
- The pre-mortem (for anything irreversible, 5 minutes before you sign)
- 4 worked composites (a crew-member decision, a channel cut, a price rise worked to the dollar, and a good decision with a bad outcome)
- 10 printable worksheets, a 32-column tracker and a 30-day setup guide (including the calibration worksheet and the quarterly scorecard)
The Problem
You do not have a decision problem. You have a speed problem, and it runs both ways. The small calls sit open for weeks. You bring them up to your spouse. You bring them up to your partner. You still do not decide.
Then a lease, a loan or a big truck shows up with a deadline attached, and you sign it in an afternoon. Too slow on the cheap calls. Too fast on the expensive ones.
And when you do decide, nobody writes down what they expected. So a quarter later there is no way to tell a good call from a lucky one.
The rail makes you faster. Only the journal makes you better.
What you need is a rail. One process every hard call runs through, aimed at the right problem, with a written expectation you can check later. That is what this kit is.
Step 0, then R.E.V.E.R.S.E.
One filter, then 7 steps. Run them in order. Stop when the answer is obvious.
0Constraint
Your business is stuck in 1 of 4 places: getting customers, selling, delivering, or running the place. A decision that does not serve that constraint gets 5 minutes or goes on the stop-doing list. You can run a perfect process on the wrong problem.
RReversible
Measured from both sides: what it costs you to undo, and whether the other person can be put back where they were. Changing a supplier for a month is reversible. Signing a 3-year lease is not. Letting someone go is not either, however cheap it is for you to hire again.
EExpected value
The weighted average across the outcomes that actually happen, in gross profit, not revenue. And the chances start from your base rate: of your last decisions like this one, how many did what you expected? The book shows one honest base rate flipping the same test from positive to negative.
VVerify
Every hard decision rides on 2 or 3 untested assumptions. Find the one that kills it if wrong, and test that first. 3 ways to test inside 24 hours. Sometimes the decision dissolves.
EEvidence
A 6-rung ladder from "somebody said it on a podcast" to "somebody paid you." A promise or a setting describes intent. Records describe behaviour. When they disagree, the records win.
RRisk
Work out your survivable-loss line: the cash you could lose this quarter without missing payroll or rent. If the worst realistic case is inside it and the expected value is positive, go. If not, shrink the bet until it fits. No upside multiple to hit.
SSecond-order
If this works exactly as planned, what new problem shows up next? Are you ready to run that one too?
EExecute
Yes, no, or a dated test. Never maybe. Then act as if it cannot be undone: write the decision document, set the check date, and leave it alone until then.
When to skip the rail
Not every decision earns it. You set your own skip line once: below this cost and this many days to undo, on both sides, you decide on the spot. We do not pick the number. A mistake that is noise to one business is a missed payroll to another.
The Decision Journal
One page per decision, written before you find out. What you expect, how confident you are out of 10, the assumption that kills it, the evidence rung, and a check date in the calendar.
On the check date you answer 3 questions separately: did I do the thing, did the thing happen, and was it given long enough. Then you grade the process and the outcome apart, because a good decision can have a bad outcome and a lucky one teaches the wrong lesson.
After a quarter, your confidence numbers become a calibration score: plain arithmetic that shows whether your gut means anything. No target from us. Your first quarter is your baseline.
Day 1, you count the decisions open in your business right now. Day 30, you count again and compare. That is the first proof it is working, or is not.
Who This Is For
An owner-operator of a small service business who keeps hires, price changes, equipment buys and people calls open for weeks, then decides the big ones fast when a deadline lands. If you have said "let me sit on it another few days" more than twice this month, this kit is written for you.
Not for you if you already log your decisions and review them on a date. Not for you if you are looking for permission to keep putting things off.
Questions
Is this legal advice for firing someone?
No. The people chapter says plainly that letting someone go is not reversible, hands the conversation to The Hard Conversations Playbook in the library, and lists the questions to ask an employment attorney first.
Does it tell me what my numbers should be?
No. Every number in the book is a round, labelled hypothetical, or a blank for yours. You set your own survivable-loss line, your own skip line and your own baseline.
Read it in your library.
This is included with The Standard, along with every other guide in the library. Your monthly re-audit tells you which one to open next. $497 to start, $297/mo to keep it running. Cancel any month and keep everything.
See The Standard →Not sure yet? Take the free Bucket Audit or tell us what's broken.
