Questioning System
Simple questions are everlasting.
- What is the true nature of reality?
- What does good looks like?
- Why are we here?
- What next?
All that I know is that I know nothing.
— Socrates
Clarity of Intent
In the age of AI it's not what you know, but knowing what and how to ask that matters most.
The problem isn't information — it's direction.
- We're drowning in content but starving for clarity
- We optimize metrics without questioning what we're optimizing for
- We solve symptoms while root causes compound
- We accept constraints that exist only in our assumptions
The quality of your questions determines:
- What problems you see (and which remain invisible)
- What solutions become possible (and which stay unimaginable)
- What you optimize for (and what you sacrifice unknowingly)
- Who you become (questions shape identity over time)
Waiho kia patai ana, he kaha ui te kaha.
VVFL Position
Questions are not a standalone practice. They are the instrument of the Reflect station in the Validated Virtuous Feedback Loop — the controller that reads the gauge and asks: is this where we should be?
CAPTURE → PRIORITIES → ATTENTION → VALUE → SYSTEMS → STANDARDS → DISTRIBUTE → REFLECT → EVOLVE
↑
Questions live here
The controller's job: read the gap between setpoint and reality, then correct. A question without a declared setpoint is noise. A question that measures the gap between where you are and where you declared you'd be is a controller output.
| VVFL Element | Applied to Questions |
|---|---|
| Setpoint | Declare what good looks like before asking |
| Gauge | Does this question change a decision? |
| Controller | The question itself — reads gap, proposes correction |
| Agent | The person or system running the loop |
Three Signals
How to know if your questions are improving — mirrors the VVFL gauge:
| Level | Signal | Proxy |
|---|---|---|
| Self | Your questions change your own decisions | Log entries in Decision Journal where the question shifted the decision |
| Transfer | Better questions out than in | The evolved question (Day 21) is unrecognizable compared to Day 1 |
| Compound | Your questions become starting points others build from | A ## Questions entry you wrote appears in someone else's decision trace |
Level 1 before Level 2. Level 2 before Level 3. Most people never leave Level 1 because they never measure.
Inner Loop
Questions feed into the tactical loop:
| Step | Practical Move | Surface |
|---|---|---|
| Questions | Clarify what matters and what is uncertain | This page |
| Problem Solving | Turn ambiguity into a defined problem statement | Problem Solving |
| Decisions | Commit to a next move and learn from outcomes | Decisions |
Perspective
Questions are not requests for information. They are invitations to transformation.
The Zeigarnik Effect
Questions trigger the Zeigarnik Effect — an open loop your mind compels you to close. This is why you:
- Binge Netflix shows (unresolved plot)
- Scroll social media endlessly (what's next?)
- Need closure in relationships (unanswered questions haunt)
Your unconscious mind continues working on open questions even when your conscious mind has moved on. This is why breakthroughs come in showers, on walks, upon waking — your background processor found something.
Harness this deliberately:
- Formulate your question with precision
- Stop working — do something unrelated
- Sleep on it
- On waking, capture raw thoughts before analysis corrupts them
The Evolution Principle
The gap between your first question and your refined question reveals the evolution of your understanding. That evolution is the insight.
Day 1: "How do I get more users?"
Day 7: "What would make users tell their friends?"
Day 14: "What transformation do users experience that they can't stop talking about?"
Day 21: "What do users become after using this?"
Track your questions over time. The trajectory matters more than any single answer.
First Principles
Before tactical questions, there are five that every conscious human must face. These are the operating system of the loop.
Context is Critical — Domain Leaves
The centre of this star is the method — types, Socratic, Five Whys, Systems Lens, Traps. The points of the star are the domain-specific application examples below. Each leaf applies the same method to a different situation; read the centre once, then come back to the leaves as situations arise.
Agency
"What do I have the agency to change?"
When feeling overwhelmed or seeking leverage
Possibilities
"What's possible here and who cares?"
When stuck in problem-solving mode
Reality
"What is the real question underneath?"
When treating symptoms, not causes
The Future
"What becomes possible without constraint X?"
When current thinking feels limited
Business
"What will machines never be able to do for you?"
When shaping strategy in an AI-accelerated world
Library
Full 22-question framework
For deep exploration and reference
Situational Wisdom
| If you're thinking... | Start with | Decision Domain |
|---|---|---|
| "What am I even optimizing for?" | Existence Questions | Decision Journal |
| "I'm overwhelmed by all the changes" | Agency Questions | Decision Algorithms |
| "I'm stuck solving the same problems" | Possibilities Questions | Tech Stack |
| "I keep treating symptoms, not causes" | Reality Questions | Decision Journal |
| "I need to find leverage in this mess" | The Systems Lens (method) | Tech Stack |
| "Current constraints limit my thinking" | Future Questions | Blockchain |
Methods
Tools for better questions. The centre of the star. Each method is a lens — choose by situation. The leaves of this section apply these methods to specific domains (existence, agency, possibilities, reality, future, business).
Question Types
Different situations need different questions. Classify before you apply.
| Type | Purpose | Example |
|---|---|---|
| Open | Expand thinking | "What do you think about...?" |
| Closed | Confirm or narrow | "Did you test this?" |
| Diagnostic | Identify the problem | "Where does it break?" |
| Upstream | Find root causes | "What caused the thing that caused this?" |
| Clarifying | Remove ambiguity | "What do you mean by...?" |
| Challenging | Test assumptions | "What's the strongest case against?" |
| Generative | Create options | "What would make this remarkable?" |
Default to open. Use closed only when you need a specific answer.
The Socratic Method
Socrates didn't teach by giving answers. He taught by asking questions that revealed what people thought they knew but didn't. The method:
- Start with genuine curiosity — not knowing the answer yourself
- Ask for definitions — "What do you mean by X?"
- Probe assumptions — "Why do you believe that?"
- Request evidence — "What would prove this wrong?"
- Explore implications — "If that's true, what follows?"
- Question the question — "Is this the right thing to be asking?"
The compressed form — Define → Probe → Challenge → Synthesise — is four doors. Rushing to the fourth means walking through the wrong one.
The Five Whys
For any surface problem, ask "why" five times:
Why are users churning?
→ They're not finding value quickly
Why aren't they finding value?
→ Onboarding doesn't show the core feature
Why doesn't onboarding show it?
→ We optimized for signup completion, not activation
Why did we optimize for signups?
→ That's what we measured
Why did we measure only signups?
→ [Root cause: metrics weren't aligned with value]
The Systems Lens
Every question asks about a system — the question is whether you see the system. The Systems Lens treats questioning as a search for leverage inside a network of relationships. Use it when you face a complex problem with many moving parts, when recurring problems resist repeated fixes, or when you need to anticipate unintended consequences.
The catalytic question: How does A relate to C, and what questions does that suggest?
Core questions:
- Feedback Loops — What are the feedback loops in this system? Which are reinforcing (amplify change) and which are balancing (stabilise)?
- Leverage Points — Where is the high-impact, low-effort intervention? Where is the bottleneck or constraint?
- Cascade Effects — How do changes cascade through the system (1st, 2nd, 3rd order)?
- Unintended Consequences — What breaks when this works?
- System Boundaries — What's inside this system, and what's outside it that still influences it?
Application process:
1. Map Components — list elements, purpose, dependencies
2. Identify Relationships — depends on / influences / constrains, with strength
3. Find Feedback Loops — reinforcing, balancing, dominant
4. Locate Leverage Points — high-impact low-effort; critical dependencies
5. Trace Cascade Effects — immediate, downstream, long-term
Worked example — Template-based code quality.
System: Code quality in a monorepo
Components: Generators → Templates → Files
Feedback Loop:
Better templates → better files → better patterns → better templates
Leverage Point: the Generator / Template layer
— affects all future files automatically
— single fix propagates everywhere
Intervention:
Fix template: 1 hour
Fix 13 instances manually: 26 hours
ROI: 2500% (25 hours saved)
Cascade:
1st order — new files use the correct pattern
2nd order — developers learn from better examples
3rd order — quality culture improves
The Systems Lens connects directly to the VVFL Position above — questions are controller outputs, the Systems Lens is how the controller finds the single intervention point that makes the loop correct.
Question Traps
Traps are questions shaped like probes but sealed shut. Learn to recognise them in others; learn to kill them in yourself.
| Trap | What It Looks Like | The Fix |
|---|---|---|
| Leading | "Don't you think we should...?" | Remove the embedded answer |
| Loaded | "Why did you fail to deliver?" | Separate the claim from the question |
| False choice | "Should we do A or B?" | Add "or something else entirely?" |
| Validation | "This is good, right?" | Ask someone who'll tell you no |
| Rhetorical | "Who would disagree with that?" | Only ask if you genuinely want the answer |
Domain Unlocking Questions
One question unlocks each domain. Learn the unlockers for the domains you work in.
| Domain | The Unlocking Question |
|---|---|
| Product | What job is the customer hiring this for? |
| Strategy | What are we saying no to? |
| Culture | What behaviour gets rewarded that we say we don't want? |
| Sales | What's the hidden objection they won't say aloud? |
| Self | What would I do if I weren't afraid? |
Question Quality Checklist
Before accepting your question as final:
- Have I challenged my key assumptions?
- Have I explored alternative framings?
- Am I guiding discovery, not imposing conclusions?
- Is this the right question to be asking?
- Am I comfortable being uncomfortable with the answer?
- Does this question open more doors than it closes?
Character
The method on this page is the know-how. The character that makes the method work lives on the Questioning capability page — intellectual humility, comfortable discomfort, patient urgency, beginner's mind, the shadow, the archetype lens.
Capability = Know-How + Character + Resources. A method without the character to run it is a checklist. Read both pages.
Confirmation
The cost of a bad answer is visible — we can see when we're wrong and correct course.
The cost of a bad question is invisible — we optimize brilliantly for the wrong thing, solve the wrong problem elegantly, build the wrong future skillfully.
Most regrets are questions unasked:
- The assumption never challenged
- The alternative never explored
- The constraint never questioned
- The "why" never pursued
The risk isn't in asking too many questions. It's in asking too few. In accepting the frame given to you. In optimizing within boundaries that don't exist.
The important thing is not to stop questioning.
— Einstein
Practice
The Question Evolution Loop
Volume alone doesn't improve question quality. Asking more of the same questions at higher frequency compounds existing blindspots. Three mechanisms actually work:
- Volume — exposure builds pattern recognition
- Reflection — examining the questions you asked (meta-cognition)
- Stakes — questions that materially change what you do next
The loop that combines all three:
FORMULATE → SLEEP → CAPTURE → REFINE → APPLY → MEASURE → EVOLVE or KILL
| Phase | When | What | Infrastructure |
|---|---|---|---|
| FORMULATE | Evening | State the question with precision before sleep | Evening ritual below |
| SLEEP | Overnight | Unconscious processing (Zeigarnik) | Biological |
| CAPTURE | Morning, raw | Thoughts before analysis filters them | Morning ritual below |
| REFINE | Session start | Is this still the right question? | Tight Five |
| APPLY | Work session | Do the work informed by the refined question | — |
| MEASURE | Session end | Did the question change the decision? | Decision Journal |
| EVOLVE or KILL | Weekly | Promote question to template — or retire it | Weekly review below |
KILL is the hardest phase. If no questions are ever retired, the library grows stale. You hold questions that once served but now just consume cognitive load. The Day 1→21 evolution example implicitly retires "how do I get more users?" by Day 14. Make it explicit.
COMPOUND is the output. The best question you evolved this week becomes the page-closing question on the most relevant /docs/ page. The ## Questions section at the bottom of every content page is where question quality compounds into the knowledge base — available to every future reader and agent.
Morning
Before starting work, ask:
- What did my unconscious work on overnight? (capture raw — no analysis yet)
- Is this still the right question, or did it shift?
- What assumption am I making that I should test today?
Evening
Before ending work, ask:
- What question did I discover today?
- Did this question change a decision? (log in Decision Journal if yes)
- What question should my unconscious work on tonight? (formulate with precision)
Weekly
- How have my questions evolved this week? Show the trajectory.
- Which questions changed decisions — and which recycled without changing anything?
- What question am I not asking that I should be?
- Retire: which questions no longer serve? Remove them from active rotation.
- Promote: which evolved question becomes a page-closing question in
/docs/?
Dig Deeper
- Existence Questions — The five questions every conscious human must face
- Questions Library — Complete 22-question framework with philosophical foundations
- Problems — What questions are trying to solve
- Problem Solving — Turn questions into action
- Decisions — Turn questions into the Perceive → Question → Act loop
- Matrix Thinking — Find connections between questions
- Prompts — Turn questions into AI conversations
- Behavioural Biases — Understand the psychology of questioning
Links
- First Principle Questions
- The Art of Learning
- Socratic Questioning Techniques
- How to Ask Better Questions
Questions
What question are you avoiding because you already know the answer would require you to change?
- If your questions haven't evolved in the last month, what are you protecting?
- Which of your current problems is actually a symptom of the wrong question?
- When did you last retire a question — and what did letting it go make possible?