Decide With Half the Facts: The Clarifier That Beats Analysis Paralysis
A high-stakes call with missing information isn't a research problem, it's a clarity problem. Here's the 4-step recipe (and the prompt) to make the call this week instead of next quarter.
The stuck moment. You run a 14-truck HVAC company. A competitor across town just put their book of business up for sale (roughly $1.1M in revenue, asking $480K). You have three weeks to put in an offer. You don't know their tech retention, how much of that revenue is one-time installs vs. recurring maintenance, or whether their best people walk the day you close. You could spend the whole window chasing answers you'll never fully get, or you could keep stalling until someone else buys it. Both feel wrong.
This is the trap: you're treating a decision problem like a research problem. More data won't save you, because the data you actually need doesn't exist yet. What you need is clarity on what's already decidable.
The recipe, the Decision Clarifier (4 steps):
- Name the real decision. Write the choice as a single sentence with a deadline: "Do I offer on this book by July 18, and at what price?" Skip vague versions like "should I grow by acquisition someday."
- Split what you know from what you're guessing. Two columns. Facts you can verify go left. Assumptions dressed up as facts go right. Most stuck decisions are 70% guesses you've been treating as knowns.
- Find the few unknowns that actually move the answer. Of your guesses, which ones, if you knew them, would flip your decision? Usually it's two or three. Ignore the rest. They're noise.
- Set a tripwire, then decide. For each decision-moving unknown, define the threshold that changes your answer ("if recurring maintenance is under 30% of revenue, I walk"). Decide now based on your best estimate, and let the tripwire tell you when to reverse.
The AI move. Paste this into Claude:
Role: You are a decision strategist who helps operators make high-stakes
calls under incomplete information. You are blunt and you do not pad.
Context: Here is the decision I'm stuck on:
[describe the decision, the deadline, the dollars at stake, and
everything you currently know. Dump it all, messy is fine]
Task: Run a Decision Clarifier on this.
1. Restate my actual decision as one sentence with a deadline.
2. Build a two-column table: VERIFIED FACTS vs. ASSUMPTIONS I'm
treating as facts. Be ruthless: flag anything I can't prove.
3. Identify the 2-3 unknowns that would actually FLIP my decision if
I knew them. Ignore unknowns that don't change the answer.
4. For each decision-flipping unknown, propose a tripwire: the
specific threshold at which my answer should change.
5. Give me your recommended call right now, based on best estimates,
and the single fastest thing I could verify before the deadline.
Format: Tables and short bullets. End with a one-line verdict.
Constraints: Don't ask me clarifying questions. Work with what I
gave you and state your assumptions. Don't hedge, and cut the fluff.
What it's worth. This turns a three-week stall into a one-hour decision, and on a $480K call, getting to a clear yes-or-no before the window closes is the difference between owning the deal and watching it go to someone faster.