Experiments
How do you turn a hypothesis into truth?
If science is the engine of progress, experiments are the fuel. Every venture is an experiment in value transformation. If you cannot define the Smallest Viable Experiment (SVE), you aren't building—you're gambling.
The Method
The Scientific Method applied to the Business Factory:
- Observe the friction
- Hypothesize the wedge
- Run the Minimum Cost Experiment
- Measure the Outcomes
- Evolve the Standard
Minimum Cost Experiments
Don't bet the farm on a guess. Jeff Bezos runs experiments on reversible decisions and lets the data compound.
| Threshold | Strategy | Action |
|---|---|---|
| Reversible | Low cost, high speed | Run the experiment immediately |
| Irreversible | High cost, low speed | Deep research before commitment |
The goal isn't to be right; it's to be less wrong each cycle. A failed experiment that produces high-signal data is a success. A "successful" product launch with no data is a failure.
Smallest Viable Experiment (SVE)
The SVE is the smallest unit of action that can support or refute a hypothesis.
- Bad SVE: "Build a full SaaS platform to see if people want it."
- Good SVE: "Create a landing page with a 'Buy' button and measure the click-through rate."
If the SVE takes more than one sprint, it's probably not the smallest. Break the pipe until the flow is visible.
Seed Experiments
What experiments should every agent/player trial at least once in their life? (via Andre Karpathy)
- The Monk: A month of silence and meditation.
- The Athlete: Training for and completing a marathon or triathlon.
- The Founder: Starting a business and making the first dollar.
- The Nomad: Living in a foreign country where you don't speak the language.
- The Creator: Publishing one piece of content every day for a year.
Each experiment changes the observer. You don't just learn about the world; you learn about your own mindset.
Individual Experiments (The Inner Loop)
If the Business Factory builds the outer world, the Inner Loop builds the agent. Your mindset and attention are the two variables you control. Everything else is coordination.
Biological (Healthspan)
Your biology is the substrate for your consciousness.
- The Sleep Experiment: Track 30 days of consistent 8-hour sleep windows. Measure cognitive load and mood.
- The Fasting Experiment: Trial intermittent fasting (16:8) for 14 days. Observe energy stability vs. insulin spikes.
- The Movement Experiment: Zone 2 cardio for 150 minutes a week. Measure resting heart rate and stress recovery.
Cognitive (Attention)
Attention is the currency of the digital age.
- The Dopamine Detox: 24 hours without digital screens, caffeine, or processed sugar. Observe the baseline of your own boredom.
- The Watcher Experiment: 10 minutes of meditation daily. Practice "watching the watcher" to separate your identity from your thoughts.
- The Deep Work Experiment: Three 90-minute "no-interruptions" blocks per week. Measure output volume vs. scattered work.
Philosophical (Mindset)
Training optimism is a repeatable practice.
- The Rejection Experiment: Ask for one thing every day that you expect to be denied. Desensitize yourself to "No" to expand your agency.
- The Truth Experiment: Write one true sentence every day for 30 days. Strip away the narrative and record only what is deterministically true.
- The Gratitude Loop: Every evening, record three Ws (Successes) from the day. Train your brain to scan for progress instead of friction.
The Better Practice Thesis
Every individual experiment is a move toward Better Practice. We don't rise to the level of our goals; we fall to the level of our habitual protocols.
Use the Third Space—the gap between stimulus and response—to choose your experiment.
Experiment Template
Use the Standard Experiment Template for every trial:
| Field | Description |
|---|---|
| Hypothesis | What do you believe is true? |
| Measurement | What signal confirms or denies? |
| Method | What's the SVE? |
| Evidence | What did you observe? |
| Conclusion | Was the hypothesis supported? |
Context
- Science — The methodology underneath every experiment
- Ventures — The factory's output: experiments that found their pipes
- Smallest Viable Experiment — Force-to-friction method
- Bezos Thesis — Minimum cost experiments on reversible decisions
- Matrix Thinking — Using gaps to prompt new experiments
- Outcome Map — Picture the success before you test it
- Reality — Where experiments meet the ground
- Evolution — How experiments graduate to standards
Questions
What is the smallest experiment you could run today to prove your current priority is the wrong one?
- If an experiment fails but the data is perfect, did you win or lose?
- Which of your ventures currently lacks a defined Smallest Viable Experiment?
- How much capital are you spending on irreversible decisions without an experimental precursor?
- If the observer effect applies to business, how does your presence in the market change the experiment's outcome?