How to Choose a Business Idea When You Have No Money
Many successful companies start with very little capital because the primary resource is expertise rather than infrastructure.
Paul Graham, co-founder of the startup accelerator Y Combinator, has repeatedly emphasised that early startups should focus on solving real problems rather than raising money. In his essay "Startups = Growth", he notes that the earliest stage of a startup is often simply a small group of people building something useful and finding users.
Read: http://www.paulgraham.com/growth.html
This idea also appears frequently in the bootstrapped startup community around Indie Hackers, where many founders describe launching businesses using existing skills and basic tools before investing significant capital.
1) Start with skills rather than products
Businesses that rely on existing skills require almost no startup capital.
Examples include:
- writing and editing
- design services
- programming
- marketing support
- research services
These businesses convert knowledge directly into value.
2) Use existing platforms
Modern infrastructure dramatically reduces startup costs.
Examples include:
- Shopify for e-commerce
- Gumroad for digital products
- Substack for newsletters
- Stripe for payments
These platforms allow entrepreneurs to launch quickly without building complex systems.
3) Validate demand before investing
Instead of building a full product, test whether people will pay.
Common validation methods include:
- pre-selling
- consulting with early users
- launching simple landing pages
Entrepreneur Eric Ries popularised this approach in The Lean Startup (2011).
Summary box
| Approach | Startup cost | Speed |
|---|---|---|
| Freelance services | very low | fast |
| Digital products | low | medium |
| E-commerce | medium | medium |
Key takeaways
- Capital is often less important than skills and initiative.
- Platforms reduce the cost of experimentation.
- Validation should happen before major investment.
Next steps
Identify one skill you already possess and test whether someone would pay for help solving a related problem.

