Find Right Question

Input: $ARGUMENTS

Interpretations

Before executing, identify which interpretation matches the user’s input:

Interpretation 1 — Unstick yourself: You’re stuck on something and suspect you’re asking the wrong question — you need to find the question that actually unblocks progress (e.g., “I keep spinning on this” or “I’m stuck and don’t know why”). Interpretation 2 — Reframe a problem: Someone gave you a problem or you’ve been working on one, but it feels off — you want to find a better framing by discovering the right question underneath the surface-level one (e.g., “something about this problem statement feels wrong” or “am I even solving the right thing?”). Interpretation 3 — Start from confusion: You’re at the very beginning of something and don’t even know what to ask yet — you need to go from vague confusion to a specific, actionable question (e.g., “I don’t know where to start” or “I’m confused about this whole area”).

If ambiguous, ask: “I can help with finding the question that unblocks you, reframing a problem you suspect is misframed, or going from confusion to a clear starting question — which fits?” If clear from context, proceed with the matching interpretation.

Purpose

When you are stuck, starting something, failing, or just confused, the bottleneck is usually not the answer — it is the question. This procedure diagnoses your actual situation and generates the most actionable question to pursue.

Depth Scaling

Default: 2x. Parse depth from $ARGUMENTS if specified (e.g., “/frq 4x [claim]”).

These are floors. Go deeper where insight is dense. Compress where it’s not.

Step 0: Why do you need a question?

Answer this honestly:

SECTION A: You’re Stuck

Step A1: Write Down What You’re Stuck On

Complete this sentence in writing:

“I am trying to _________ but I can’t because _________.”

You should see: A concrete statement with two parts — a goal and an obstacle.

If you can’t complete the sentence, your problem isn’t that you’re stuck. You don’t have a clear goal. Go to SECTION B.

Step A2: Check the Obstacle

Look at the “because” part of your sentence. Is the obstacle:

(a) A lack of information? (“I don’t know how to…”) Your question is: “Where can I find [specific info]?” This is an operational question. It’s probably the right one. Go find the information. DONE.

(b) A lack of options? (“I can’t think of a way to…”) Your question is: “What approaches exist that I haven’t considered?” Go to Step A3.

(c) A conflict between options? (“I can’t decide between…”) Your question is: “What criteria would make this decision obvious?” Go to Step A4.

(d) A feeling? (“I’m afraid to…” / “I don’t want to…” / “It feels wrong…”) Your question is NOT about the task. It’s about you. Go to Step A5.

(e) Something external? (“They won’t let me…” / “I don’t have the resources…”) Your question is: “Is this constraint real, or assumed? And if real, what changes?” Go to Step A6.

Step A3: Expand Your Option Space

List every approach you’ve already considered. Then for each one, ask:

• “What’s the OPPOSITE of this approach?”
• “What would a complete beginner try?”
• “What would someone with unlimited resources do?”
• “What would happen if I did nothing?”

Your question is the one that points at the most promising unexplored direction.

Test your question with SECTION F.

Step A4: Clarify Your Criteria

For each option you’re deciding between, write:

“If I choose this, I get _________ but lose _________.”

Now look at what you’d gain and lose. Ask:

“Which of these gains/losses matters most to me in [your time horizon — 1 month? 1 year? 5 years?]?”

Your question is: “What do I actually value more: [gain A] or [gain B]?”

Test your question with SECTION F.

Step A5: Address the Feeling

Name the feeling. Write it down. Then ask: “What is this feeling protecting me from?”

Common patterns:

• Fear: “What’s the worst realistic outcome, and can I handle it?”
• Overwhelm: “What is the ONE next thing I need to do?”
• Perfectionism: “What would ‘good enough’ look like?”
• Boredom: “What about this have I outgrown?”
• Resentment: “What did I agree to that I shouldn’t have?”

Your question is the one above that fits your feeling.

Test your question with SECTION F.

Step A6: Test the Constraint

Write down the constraint. Then answer:

“Has anyone ever done what I want despite this constraint?”

• Yes: Your question is: “How did they do it?”
• No: “Is there a different goal that avoids this constraint entirely?”
• Unsure: “What would I need to check to find out?”

Test your question with SECTION F.

SECTION B: You’re Starting Something New

Step B1: Define “Done”

Complete this sentence:

“This is successful when _________.”

If you can’t complete the sentence, your question is: “What would success look like, concretely?” Answer THAT question first, then return to B1.

Step B2: Find the Riskiest Assumption

List everything that must be true for your “done” definition to be achievable. These are your assumptions.

Rank them: Which is LEAST certain?

Your question is: “Is [least certain assumption] actually true? How would I find out quickly?”

Test your question with SECTION F.

SECTION C: Something Isn’t Working

Step C1: Specify the Failure

Complete this sentence:

“I expected _________ but instead I got _________.”

If you can’t specify what you expected, you didn’t have a clear goal. Go to SECTION B.

Step C2: Locate the Gap

The gap between expected and actual has a cause. List three possible causes, then for each ask: “What evidence would confirm or rule this out?”

Your question is: “Which of [cause 1 / cause 2 / cause 3] is actually responsible, and how can I test it fastest?”

If you can only think of one cause — WARNING: You’re probably anchored on the obvious explanation. Ask: “What if the obvious cause is wrong? What ELSE could produce this result?”

Test your question with SECTION F.

SECTION D: You Want to Understand Something

Step D1: Check Your Motive

Ask yourself honestly: “If I understood this, what would I DO differently?”

• Clear action: Your question is about HOW to do that action. Go to SECTION A, Step A2a.
• No clear action: Ask: “Is this curiosity or avoidance?” Be honest.
  • Curiosity: Go to Step D2.
  • Avoidance: You’re procrastinating. What are you avoiding? Go to SECTION A, Step A5.

Step D2: Find the Specific Confusion

Write down what you think you understand. Then mark the point where your understanding breaks down.

Your question lives at THAT boundary — the edge between what you know and what you don’t.

Frame it as: “I understand [X] but I don’t understand [Y]. What connects X to Y?”

Test your question with SECTION F.

SECTION E: Someone Gave You a Problem

Step E1: Check the Frame

Write down the problem as given to you. Then ask three questions about it:

1. “Is this the real problem, or a symptom?” To test: Ask “Why is this a problem?” If the answer points to a DIFFERENT problem, follow that chain until you hit a root.

2. “Whose problem is this really?” If it’s not yours, your question is: “What’s my actual role here?” (solver, advisor, supporter, affected party?)

3. “Has the problem-giver already decided the answer?” If yes, they don’t want a question — they want validation. Your question is: “Do I agree with their proposed solution?” If yes, help execute. If no, raise your concern.

Step E2: Reframe If Needed

If Step E1 revealed a different problem, write the NEW problem statement and go to the appropriate section (A through D).

If the problem survived E1 unchanged, it’s well-framed. Go to the appropriate section based on what type of problem it is.

SECTION F: Validate Your Question

Step F1: The Five Tests

Apply each test to your candidate question. Score: Pass (1) or Fail (0).

Test 1 — ACTIONABILITY: “If I got a clear answer to this question, would it change what I do?” Yes = Pass / No = Fail

Test 2 — SPECIFICITY: “Could two different people interpret this question differently and give different (valid) answers?” No = Pass / Yes = Fail (too vague, narrow it)

Test 3 — HONESTY: “Am I asking this to learn, or to confirm what I already believe?” To learn = Pass / To confirm = Fail (find what you’re avoiding)

Test 4 — LEVEL: “Is this question at the right level? Could there be a deeper question underneath it?” Right level = Pass / Deeper question exists = Fail (ask the deeper one)

Test 5 — ANSWERABILITY: “Is there a way — some observation, experiment, or analysis — that could answer this question?” Yes = Pass / No = Fail (reframe to be answerable)

Scoring:

• 5/5: You have your question. Go pursue the answer.
• 4/5: Fix the failing test and recheck.
• 3/5 or below: Go back to your section and dig deeper. You’re still on the surface.

Step F2: Final Formulation

Write your validated question in this format:

QUESTION: [your question] WHY THIS QUESTION: [1 sentence on why this is the right thing to ask right now] WHAT A GOOD ANSWER LOOKS LIKE: [what form the answer would take] FIRST STEP TO ANSWER IT: [one concrete action]

Quick Reference Cards

Card 1: Question-Level Ladder

• If you’re asking “How?” — check if you should ask “Should I?”
• If you’re asking “Should I?” — check if you should ask “What for?”
• If you’re asking “What for?” — check if you should ask “What matters?”
• Always go one level higher than feels natural.

Card 2: Red Flags That You Have the Wrong Question

• You already know the answer you want
• The question has been asked before and the answer didn’t help
• The question makes you feel smart rather than uncertain
• Everyone around you is asking the same question
• The question is about blame rather than mechanism

Card 3: Emergency Shortcut When completely lost, ask ONE of these:

• “What would I do if I couldn’t fail?”
• “What am I avoiding?”
• “What’s the simplest version of this?”
• “What would change my mind?”
• “What do I know for sure?”

Common Mistakes

1. Starting with “How?” instead of “What?” or “Why?” — How-questions assume the goal is right. Check the goal first.
2. Asking the question everyone else is asking — Consensus questions produce consensus answers. The right question is often the one nobody is asking.
3. Refusing to ask simple questions — “What are we actually trying to do?” is the most powerful question in most situations, and the one people are most embarrassed to ask.
4. Treating question-finding as a solo activity — Explaining your problem to someone else often reveals the question faster than thinking alone. The question emerges in the gap between your understanding and theirs.
5. Asking questions in the wrong order — Foundational questions first, operational questions last. “Should I do this?” before “How should I do this?”
6. Optimizing the question instead of testing it — A mediocre question that gets answered teaches you more than a perfect question that stays theoretical.

When to Override This Procedure

• Emergency / time pressure — Skip to the Emergency Shortcut (Card 3). Any question is better than no question when you need to act NOW.
• You already know the question but are stalling — Sometimes you know what to ask and you’re using “finding the right question” as procrastination. If a question came to mind immediately and makes you uncomfortable, that’s probably it.
• The situation is genuinely novel — No procedure can navigate a truly unprecedented situation. Use the procedure to get started, then trust your judgment as new information arrives.
• You’re in a collaborative context — Finding the right question WITH others is a different process (more dialogic, less procedural). Use this procedure for your own thinking, then bring your best question to the group.