Sam Altman Startup Board Guide

Build a Claude Project that reviews your startup using Sam Altman’s public startup principles as an evaluation framework.

Brand: Life of Arjav

Instagram: @lifeofarjav

---

What this guide helps you build

This guide helps you create a Claude Project called Startup Board.

The goal is not to make Claude pretend to be Sam Altman. That is the wrong framing.

The goal is to give Claude:

1. A source pack of Sam Altman’s public startup writing
2. Your real business context
3. Clear review instructions
4. Repeatable board meeting prompts

Then Claude can review your startup with more context, more discipline, and less generic advice.

Use this as a thinking system, not as a replacement for customers, operators, investors, or your own judgment.

---

The core idea

Most founders ask AI vague questions like:

What should I do to grow my startup?

That produces vague advice.

A better setup is:

Use these startup principles as your evaluation framework. Use my real business files as evidence. Separate facts from guesses. Challenge weak assumptions. Tell me what to do next.

The difference is massive.

Bad AI setup gives you motivation.

Good AI setup gives you a mirror.

---

Source pack

Add these public Sam Altman resources to your Claude Project.

1. Startup Playbook

Link: https://playbook.samaltman.com/

Why it matters:

This is the foundation. It covers the core startup pillars:

• Idea
• Market
• Product
• Team
• Execution

Use it as the main evaluation framework.

---

2. Startup Advice

Link: https://blog.samaltman.com/startup-advice

Why it matters:

This is a short list of direct startup advice. It is useful for founder discipline, execution, focus, product quality, team quality, growth, and resourcefulness.

Use it to pressure test whether your current priorities are actually the priorities that matter.

---

3. Startup Advice, Briefly

Link: https://blog.samaltman.com/startup-advice-briefly

Why it matters:

This is useful for early stage founders because it emphasizes starting with an idea or project, not rushing into the performance of running a company.

Use it when you are still shaping the idea, the product, or the first narrow market.

---

4. How To Be Successful

Link: https://blog.samaltman.com/how-to-be-successful

Why it matters:

This is not only about startups, but it is useful for founder behavior.

Use it to review:

• Focus
• Sales ability
• Independent thinking
• Risk taking
• Ambition
• Work ethic
• Long term compounding

---

5. Hard Startups

Link: https://blog.samaltman.com/hard-startups

Why it matters:

This is useful if your company is ambitious, technical, infrastructure heavy, regulated, deep tech, AI native, or unusually difficult to build.

Use it to ask whether the difficulty of your startup creates a moat or just creates pain.

---

6. How Things Get Done

Link: https://blog.samaltman.com/how-things-get-done

Why it matters:

This is useful for execution review.

Use it to check whether you are focused on the work that actually moves the company forward or hiding inside fake work.

---

How to create the Claude Project

Step 1: Create a new project

Open Claude.

Create a new Project.

Name it:

Startup Board

Optional description:

A startup review system that uses public startup principles and my business context to help me make better founder decisions.

---

Step 2: Upload the source pack

Upload the public startup material listed above.

Suggested folder name:

01 Sam Altman Principles

Suggested file names:

startup-playbook.md
startup-advice.md
startup-advice-briefly.md
how-to-be-successful.md
hard-startups.md
how-things-get-done.md

Keep names simple. Claude should be able to understand what every file is without guessing.

---

Step 3: Upload your real business context

Create a second folder called:

02 My Business

Upload these files.

product-overview.md
ideal-customer-profile.md
pricing-model.md
current-metrics.md
customer-feedback.md
current-priorities.md

Your AI board is only as good as the context you give it.

No business context means generic advice.

Generic advice is cheap. Useful judgment requires evidence.

---

Business document templates

Use these templates if you do not already have clean business docs.

---

Product Overview Template

# Product Overview

## Product name

[Write the product name]

## One sentence description

[Explain what the product does in one simple sentence]

## Who it is for

[Describe the specific customer]

## Problem it solves

[Describe the urgent problem]

## Current product status

Choose one:

-Idea
-Prototype
-MVP
-Live product
-Revenue generating
-Scaling

## Core features

1.[Feature 1]
2.[Feature 2]
3.[Feature 3]

## What users can do today

[Describe what is actually working right now]

## What is still missing

[Describe gaps honestly]

## Main product risk

[What could make this product fail?]

---

Ideal Customer Profile Template

# Ideal Customer Profile

## Best fit customer

[Describe the exact person or company that gets the most value]

## Industry

[Industry or niche]

## Company size

[Revenue, employees, team size, or stage]

## Buyer role

[Who makes the decision?]

## User role

[Who actually uses the product?]

## Pain points

1.[Pain point 1]
2.[Pain point 2]
3.[Pain point 3]

## Trigger events

What happens right before they need this?

1.[Trigger 1]
2.[Trigger 2]
3.[Trigger 3]

## Bad fit customers

Who should we avoid?

1.[Bad fit 1]
2.[Bad fit 2]
3.[Bad fit 3]

## Proof we understand the customer

Include real quotes, call notes, emails, or objections.

---

Pricing Model Template

# Pricing Model

## Current pricing

[Explain your current pricing]

## Packaging

[Explain what is included]

## Why this pricing makes sense

[Explain your reasoning]

## Customer willingness to pay evidence

[Include proof from sales calls, closed deals, objections, or usage]

## Main pricing objection

[What do prospects push back on?]

## Unit economics

[Include gross margin, cost to serve, delivery time, support cost, or payback period if relevant]

## Pricing experiments to test

1.[Experiment 1]
2.[Experiment 2]
3.[Experiment 3]

---

Current Metrics Template

# Current Metrics

## Time period

[Example: Last 30 days]

## Acquisition

-Website visitors:
-Leads:
-Demos booked:
-Demos completed:
-Trials started:
-Customers closed:

## Conversion

-Visitor to lead:
-Lead to demo:
-Demo to close:
-Trial to paid:

## Revenue

-MRR:
-ARR:
-Cash collected:
-Average contract value:
-Churn:
-Expansion:

## Product usage

-Active users:
-Weekly active users:
-Retention:
-Core action completed:
-Time to value:

## Sales activity

-Outreach sent:
-Replies:
-Positive replies:
-Calls booked:
-Proposals sent:
-Deals closed:

## Biggest metric concern

[What number worries you most?]

## Metric you trust least

[Which number may be misleading?]

---

Customer Feedback Template

# Customer Feedback

## Best customer quotes

1."[Quote]"
2."[Quote]"
3."[Quote]"

## Common objections

1.[Objection 1]
2.[Objection 2]
3.[Objection 3]

## Feature requests

1.[Request 1]
2.[Request 2]
3.[Request 3]

## Complaints

1.[Complaint 1]
2.[Complaint 2]
3.[Complaint 3]

## What customers already love

[Describe what gets repeated praise]

## What customers ignore

[Describe what you built that customers do not care about]

## Evidence of urgency

[Show proof that customers need this now, not someday]

---

Current Priorities Template

# Current Priorities

## Current top 3 priorities

1.[Priority 1]
2.[Priority 2]
3.[Priority 3]

## Why these are the priorities

[Explain the logic]

## What we are saying no to

1.[No 1]
2.[No 2]
3.[No 3]

## Main constraint

Choose one:

-Distribution
-Product quality
-Customer urgency
-Team capacity
-Cash
-Focus
-Technical risk
-Market timing

## Biggest unresolved decision

[What decision are you avoiding?]

## What we need to learn next

[What evidence would make the path clearer?]

---

Claude Project instructions

Paste this into your Claude Project instructions.

You are helping me review my startup using the uploaded public startup principles as an evaluation framework.

Important rules:

1. Do not imitate Sam Altman.
2. Do not claim to speak for Sam Altman.
3. Do not invent his opinions.
4. Use the uploaded source principles behind important recommendations.
5. Use my business files as evidence.
6. Separate facts, guesses, and recommendations.
7. Challenge weak assumptions directly.
8. Be specific, practical, and honest.
9. If evidence is missing, say what evidence is missing.
10. Do not give generic startup advice unless it clearly applies to my context.

When reviewing my company, evaluate:

- Problem importance
- Customer urgency
- Market quality
- Product usefulness
- User evidence
- Distribution strategy
- Pricing logic
- Founder focus
- Execution speed
- Team gaps
- Fundraising readiness if relevant

For every major recommendation, include:

- The principle behind it
- The evidence from my business files
- The risk if ignored
- The next action
- What evidence would change the recommendation

Default output format:

1. Executive summary
2. What is working
3. What is weak
4. What I am avoiding
5. Biggest strategic risk
6. Highest leverage next move
7. 7 day action plan
8. Evidence needed next
9. Questions I should answer before the next review

---

First review prompt

Use this after uploading your source pack and business context.

Review my startup like a serious startup board.

Use the uploaded startup principles as your evaluation framework.
Use my business files as evidence.

Do not imitate Sam Altman or invent his opinions.
Do not flatter me.
Do not give generic advice.

Tell me:

1. What is working
2. What is weak
3. What I am avoiding
4. Where my assumptions are unsupported
5. What my biggest risk is
6. What I should do next
7. What evidence would change your recommendation

Include a scorecard from 1 to 10 for:

- Problem importance
- Customer urgency
- Market quality
- Product usefulness
- User evidence
- Growth engine
- Pricing clarity
- Founder focus
- Execution speed

Then give me a 7 day action plan.

---

Weekly board meeting prompt

Use this once per week.

Run my weekly Startup Board review.

Compare my current priorities, metrics, customer feedback, and execution against the uploaded startup principles.

Focus on what changed since the last review.

Output:

1. This week's headline
2. What improved
3. What got worse
4. What stayed unclear
5. The most important decision I am avoiding
6. The one metric I should care about most right now
7. The work I should stop doing
8. The work I should double down on
9. The highest leverage next 7 days
10. Evidence to collect before the next review

Be direct. If I am confusing motion with progress, say so.

---

Customer evidence review prompt

Use this after sales calls, demos, interviews, support chats, or onboarding calls.

Review this customer feedback.

Use the uploaded startup principles and my business context.

Tell me:

1. What customers clearly care about
2. What customers are politely ignoring
3. What pain seems urgent
4. What pain seems weak
5. Which objections matter
6. Which objections are excuses
7. What product changes are justified by evidence
8. What product changes would be premature
9. What I should ask in the next 10 customer conversations

Separate real evidence from my interpretation.

---

Feature prioritization prompt

Use this before building new features.

Review these proposed features against my startup context.

For each feature, score:

- Customer urgency
- Strategic importance
- Revenue impact
- Speed to ship
- Evidence strength
- Risk of distraction

Then classify each feature as:

- Build now
- Test manually first
- Defer
- Kill

Be strict. Assume building the wrong thing is expensive.
Tell me which feature I am emotionally attached to but should probably not build.

---

Fundraising readiness prompt

Use this only if fundraising is relevant.

Review whether this startup is ready to fundraise.

Use the uploaded startup principles and my current business files.

Evaluate:

1. Strength of idea
2. Market quality
3. Product traction
4. Founder market fit
5. Growth evidence
6. Narrative clarity
7. Investor risk
8. What questions investors will ask
9. What answers are currently weak
10. What must improve before fundraising

Give me:

- A fundraising readiness score
- The strongest investor story
- The weakest part of the pitch
- A list of missing proof points
- A 30 day plan to become more fundable

---

The scorecard

Use this scorecard when reviewing your startup.

---

How to interpret the review

Do not treat Claude’s answer as truth.

Treat it as a structured advisory memo.

The value comes from:

• Forcing your context into one place
• Making your assumptions visible
• Comparing your actions against strong principles
• Creating a weekly operating rhythm
• Turning vague founder anxiety into specific next actions

The output is useful only if your inputs are honest.

If you hide weak metrics, vague positioning, customer objections, or failed experiments, the review will be fake.

Do not use AI to protect your ego.

Use it to compress learning cycles.

---

Recommended folder structure

Startup Board/
  01 Sam Altman Principles/
    startup-playbook.md
    startup-advice.md
    startup-advice-briefly.md
    how-to-be-successful.md
    hard-startups.md
    how-things-get-done.md

  02 My Business/
    product-overview.md
    ideal-customer-profile.md
    pricing-model.md
    current-metrics.md
    customer-feedback.md
    current-priorities.md

  03 Reviews/
    weekly-board-review.md
    customer-evidence-review.md
    feature-priority-review.md
    fundraising-readiness-review.md

  04 Decisions/
    decision-log.md
    assumptions-log.md
    experiment-log.md

---

Decision log template

# Decision Log

## Decision

[What did we decide?]

## Date

[Date]

## Why we made this decision

[Reasoning]

## Evidence used

1.[Evidence 1]
2.[Evidence 2]
3.[Evidence 3]

## Alternatives considered

1.[Alternative 1]
2.[Alternative 2]

## Risk

[What could go wrong?]

## Review date

[When will we revisit this?]

---

Assumptions log template

# Assumptions Log

| Assumption | Confidence | Evidence | Test | Deadline |
|---|---:|---|---|---|
| Customers urgently need this | 5/10 | 3 calls mention it | Interview 10 prospects | Friday |
| They will pay $X | 3/10 | No paid proof yet | Send paid offer | Tuesday |
| This market is reachable | 4/10 | Early outbound data | Test 3 segments | Next week |

---

Experiment log template

# Experiment Log

## Experiment

[What are we testing?]

## Hypothesis

[What do we believe will happen?]

## Success metric

[What number proves it worked?]

## Timeframe

[Start date to end date]

## Result

[What happened?]

## Learning

[What did we learn?]

## Decision

Choose one:

-Continue
-Change
-Stop

---

The brutal operating rule

If your Startup Board keeps giving generic advice, the problem is usually not Claude.

The problem is your inputs.

You probably gave it:

• Weak context
• Vague metrics
• No customer evidence
• No decision history
• No honest constraints

Better inputs create better reviews.

Founder clarity in, founder clarity out.

---

Copy and paste quick start

Use this when you want the fastest possible setup.

I am building a Claude Project called Startup Board.

Your job is to review my startup using the uploaded startup principles as the evaluation framework and my business files as evidence.

Do not imitate Sam Altman.
Do not invent opinions.
Separate facts from guesses.
Challenge weak assumptions.
Tell me what is working, what is weak, what I am avoiding, and what I should do next.

For every recommendation, include:
1. The principle behind it
2. The evidence from my files
3. The risk if ignored
4. The next action
5. What evidence would change your mind

---

Final note

This system will not save a bad startup.

But it can help you see the truth faster.

That is the whole point.

Build the board.

Feed it real evidence.

Review weekly.

Act on what hurts.

---

Life of Arjav

Follow @lifeofarjav for practical AI and startup systems.

Comment SAM on the Instagram post to get the Startup Board guide.