The Decision Playbook: make hard calls fast — and stop relitigating them

Here’s an audit worth running: list everything on your plate that has stalled, and next to each item write what it’s actually waiting on. Rarely is it effort, money, or skill. It’s a decision you haven’t made — which vendor, which offer, whether to fire the client, whether to kill the project. Unmade decisions don’t sit quietly, either. Each one bills you daily: it resurfaces in the shower, hijacks the first twenty minutes of deep work, and forces every adjacent choice to stay provisional. Call it decision debt, and it compounds like the financial kind.

The fix is not “become smarter” or “trust your gut.” It’s procedural. Fast, calm deciders aren’t more confident than you — they classify decisions before they make them, so each one gets exactly the deliberation it deserves and not a minute more. Then they commit in a way that makes reopening the question expensive. That second half matters as much as the first: a decision you keep relitigating was never made, just postponed with paperwork.

Step 1: Sort by door type (2 minutes)

Before weighing any options, ask one question: what would it cost to undo this in 30 days?

• Two-way doors — cheap to reverse. Tools, pricing experiments, content topics, meeting formats, most hires of contractors, almost every purchase under a week’s income. If it goes wrong, you walk back through the door. These deserve minutes, not meetings.
• One-way doors — expensive or impossible to reverse. Quitting a job, signing a lease or a long contract, taking investment, a public commitment, anything involving your reputation or a relationship you can’t replace. These deserve the full playbook below.

Most people run one process for everything — so they agonize over two-way doors and, exhausted, rush the one-way doors. Flip it. The honest ratio for a working professional is roughly 20 two-way decisions for every one-way door. If everything feels irreversible, that’s not prudence; that’s anxiety doing your sorting. The same test drives what you should delegate to AI: reversible plus checkable means hand it off.

Step 2: Size the research to the stakes

For one-way doors, gather information — but put a meter on it. The working rule used by people who decide for a living: act around 70% of the information you wish you had. At 50% you’re gambling; waiting for 90% costs more in delay than the extra certainty saves in error, because the situation moves while you study it.

Operationalize it with an information deadline, set when you start, not when you’re tired: “I decide Friday at noon with whatever I have.” Then define what would actually change your mind — write down the two or three facts that would flip the call, and chase only those. Research that can’t change the decision is procrastination wearing a lab coat.

Step 3: Run the 10-minute pre-mortem

Before committing to a one-way door, set a timer and write from this prompt: “It’s six months from now and this decision failed badly. What happened?” Force out five distinct causes. Prospective hindsight surfaces risks that “pros and cons” lists miss, because you’re explaining a failure instead of defending a preference.

Then mark each cause: M for mitigable-now (cheap insurance exists — buy it), A for acceptable (survivable, priced in), or D for dealbreaker (if you can’t mitigate it, this is your real answer). A pre-mortem that produces zero dealbreakers and two cheap mitigations is a green light with the seatbelt already fastened.

Step 4: Commit with a kill criterion

This is the step that ends relitigation. A bare decision (“we’re doing X”) invites re-argument every time X has a bad week. A committed decision has a tripwire built in:

• The format: “We do X until [metric] falls below [number] or until [date], whichever comes first. Before that, we don’t reopen it.”
• The announcement script: “Decision’s made: we’re doing X. We looked hard at A and B — B lost on cost, A on timing. We review on [date] or if [tripwire]. Until then I need everyone rowing, including the people who preferred B.”
• The disagree-and-commit script (for when you’re the overruled one): “I’d have chosen B and I’ve said why. The call is made, so I’m executing A at full effort — and I’ll flag it if we hit the tripwire.”

The kill criterion does two jobs at once: it makes commitment psychologically safe (you’re not married to the decision, only to the review date), and it converts second-guessing from a mood into a measurement. New discomfort is not new information. Only the tripwire reopens the case.

Step 5: Log it in five lines

Every one-way door and any two-way door you argued about for more than a day goes in a decision journal. The entry takes ninety seconds:

1. Date + decision: what you chose, in one sentence.
2. Why: the top two reasons, as known today.
3. Expected outcome: what you think happens, with a rough probability (“70% this pays for itself by March”).
4. Kill criterion: the tripwire and review date.
5. State: one word on how you felt deciding (rested, rushed, cornered).

Review the journal monthly. Two payoffs: calibration — you learn whether your 70%s are really 70%s — and armor against hindsight. When a good decision has a bad outcome, the journal proves it was still the right call with the information you had. That distinction, decision quality versus outcome quality, is the single biggest upgrade most people can make to how they judge themselves.

Worked example: the client you should probably fire

Freelancer, five clients. One pays $2,500/month — 30% of revenue — and generates 60% of the stress: scope creep, slow payment, Sunday emails. The decision has been “pending” for four months.

• Door type: feels one-way (“I can’t un-fire them”), but run the 30-day test: could you replace $2,500/month within a quarter? Pipeline says probably. Call it a heavy two-way door — deserves a week, not a season.
• 70% check: the only fact that would flip the call is whether two warm leads convert. Information deadline: end of month.
• Pre-mortem: failure causes — leads don’t convert (M: keep client until one signs), client badmouths you (M: offer a 30-day handoff, leave clean), cash dips (A: three months of runway covers it). Zero dealbreakers.
• Commitment: “If one lead signs by the 31st, I give notice with a 30-day transition. Tripwire to reverse course: pipeline drops below two active conversations.”
• Journal: logged, 75% confident this raises effective hourly rate within one quarter. Review September 1.

Four months of low-grade dread, converted into a two-week process with an exit ramp. That’s the playbook working.

Failure modes (and the patch for each)

• Research theater. Treating a two-way door like a dissertation. Patch: the 30-day undo test, out loud, before any spreadsheet gets opened.
• Consensus shopping. Asking advisors one at a time until someone endorses the option you already wanted. Patch: pick your two advisors up front, ask both the same question, then decide alone.
• The disguised non-decision. “Still gathering info” with no decision date is a decision — for the status quo, made daily, by default. Patch: every open question gets an information deadline the moment it’s raised.
• Relitigating on discomfort. Week two feels bad, so the debate reopens — with zero new facts. Patch: point at the kill criterion. Not tripped? Not reopened.
• Moving the goalposts. Setting the tripwire after results arrive, so nothing ever triggers. Patch: the kill criterion is written in the journal before execution starts, or it doesn’t count.
• Deciding depleted. One-way doors at 11 PM after a brutal day go badly — state is part of the input, which is why the journal tracks it. Patch: big calls get made before noon, after sleep, never mid-spiral. Pressure has its own playbook; use it first.

The bottom line: your backlog isn’t a to-do list problem, it’s a decision queue. Sort every item by door type this week: two-way doors get decided today at 70% information, one-way doors get the pre-mortem, a kill criterion, and five lines in the journal. Then close the case file. The energy you’ve been spending on re-deciding is the energy your actual goals have been waiting for.