How to Start a Startup
The complete founder's playbook — from validating your idea to raising your Series A. With step-by-step checklists, free tools, and links to everything you need.
Why this matters
90% of startups fail, and the #1 reason is building something the market doesn't need. Idea validation costs almost nothing compared to building a product that nobody buys. The best founders treat their idea as a hypothesis — not a conclusion — and design cheap experiments to test it before committing months of their life.
What idea validation actually means
Validation isn't asking your friends if your idea is good — they'll say yes to be nice. Real validation means finding evidence that strangers will pay for what you're building. This can take many forms: pre-orders, letters of intent, waitlist signups, landing page conversion rates, or conversations with 50+ potential customers who describe the exact pain point you're solving. The goal is to move from "I think this is a good idea" to "I have data showing people want this."
The Mom Test
The most important framework for customer discovery is the Mom Test (from Rob Fitzpatrick's book). The core principle: never ask people if your idea is good. Instead, ask about their life, their problems, and their current workflows. "Would you use an app that does X?" is a terrible question — everyone says yes. "How do you currently handle X? What have you tried? How much does this problem cost you?" These questions reveal real behavior, not hypothetical intentions.
Minimum Viable Experiment
Before building an MVP, run a Minimum Viable Experiment. Create a landing page that describes your product with a waitlist signup or "Buy Now" button. Run $500 of targeted ads. If you can't get a 5% conversion rate on a landing page, you have a messaging problem or a demand problem. Either way, it's better to find out now. Other MVEs: a Typeform survey sent to 100 target users, a concierge version where you manually deliver the service, or a "Wizard of Oz" prototype that looks automated but isn't.
Evaluating market size
A great product in a tiny market still fails. Use TAM (Total Addressable Market), SAM (Serviceable Addressable Market), and SOM (Serviceable Obtainable Market) to estimate whether your market is big enough to build a venture-scale business. VCs typically want to see a $1B+ TAM for venture-scale returns. But be specific — "the global healthcare market is $8 trillion" is meaningless. What's the actual slice you can capture?
Real-world example
Dropbox validated their idea with a 3-minute explainer video before writing any code. Drew Houston posted it to Hacker News, and signups went from 5,000 to 75,000 overnight. That single video validated both demand and willingness to try the product — all before the product existed.
Checklist
0/8 completeReady to model your startup's economics?
Use our free interactive tools to model dilution, runway, SAFE conversions, cap tables, unit economics, and more.