Unknown

Unknown: Strategy Blind Spot Analysis

Surface hidden assumptions and strategic blind spots using the Known/Unknown 4-quadrant lens.

When to Use

• Strategy documents that look coherent but are hard to execute
• Plans with unclear risk boundaries
• Decision discussions where the team asks: "What are we missing?"

For requirement clarification, use vague. For content/form leverage decisions, use metamedium.

Core Lens

Classify insights into four quadrants:

• Known Knowns (KK): validated facts and working mechanisms
• Known Unknowns (KU): explicit uncertainties requiring experiments
• Unknown Knowns (UK): existing but underused assets/capabilities
• Unknown Unknowns (UU): unobserved risks requiring early-warning antennas

Default resource split (adjustable):

• KK 60%: systematize
• KU 25%: experiment
• UK 10%: leverage
• UU 5%: install antennas

Non-Negotiable Interaction Rule

Use AskUserQuestion for each round (R1/R2/R3). Ask with hypothesis options, not open free-text prompts.

3-Round Pattern

• R1 (3-4 questions): broad scan across all quadrants
• R2 (2-3 questions): drill into weakest link discovered in R1
• R3 (2-3 questions, optional): execution detail for top priorities

Cap total questions at 7-10.

Protocol

Phase 1: Intake

If a doc is provided, extract goals, assumptions, dependencies, and unstated constraints.

Phase 2: Context Sweep

Scan project context (README, plans, prior notes) to identify likely UK candidates (assets the team has but is not using).

Phase 3: Draft + R1

Create a provisional 4-quadrant draft, then run R1 AskUserQuestion (single batched call).

Phase 4: R2 Deepening

Design R2 only from R1 responses:

• Compound bottlenecks
• Surprising answers
• "Other" selections

Phase 5: R3 Execution (optional)

If needed, confirm owner, timeline, stop-criteria, and promotion conditions.

Phase 6: Playbook Output

Produce a structured playbook using references/playbook-template.md.

Rules

1. Hypotheses as options 2. R2 derives from R1, R3 derives from R2 3. Include both what to start and what to stop 4. Every KU gets promotion condition and kill condition 5. Prefer small experiments over large speculative plans

Additional Resources

• references/question-design.md
• references/playbook-template.md

{Topic}: Known/Unknown Quadrant Playbook

Current State Diagnosis

Quadrant Matrix

1. Known Knowns (Systematize)

2. Known Unknowns (Experiment)

For each KU:

• Hypothesis
• Experiment
• Success criteria
• Deadline
• Promotion condition
• Kill condition

3. Unknown Knowns (Leverage)

4. Unknown Unknowns (Antennas)

Strategic Decision: What to Stop

Execution Roadmap (Week-by-Week)

Core Decision Principles (3-5)

Unknown Question Design Guide

AskUserQuestion Formatting

• Batch up to 4 questions per call
• Prefer 3-4 options per question
• Use multiSelect: true when causes can be combined
• Keep options mutually distinguishable

Round-Specific Question Types

R1 (Broad Scan)

1. Certainty stress-test (KK boundary) 2. Weakest link selection (KU focus) 3. Underused asset detection (UK discovery) 4. Biggest fear/risk scenario (UU probe)

R2 (Weak Link Drilldown)

• Root cause disambiguation
• Constraint ranking
• Owner/capability gap checks

R3 (Execution Detail)

• First experiment design
• Timebox/deadline selection
• Promotion or kill criteria

Anti-Patterns

• Open questions without options
• 5+ options with overlapping meaning
• Repeating prepared questions regardless of prior answers