How to start an online business: market, testing, and mindset

Executive overview

Most early-stage founders spend money building before they know anyone wants what they're making. The fix is to test with a landing page before writing a line of code.

Choose a market you're passionate about — but only pursue it if the wave is growing. Validate quickly, manually, and cheaply. Only scale what already works.

The worst outcome of a failed startup is a bruised ego, not ruin — so test early and often.

Choosing the right market

• Ask three questions before committing: Am I passionate about this problem? Is the market growing? Would I still work on it when everything crashes?
• Passion alone isn't enough — a shrinking market limits impact regardless of effort
• The wave matters more than the rider: the same effort on an exploding market delivers far more value
• LinguaTrip taught this lesson: a study-abroad language business hit hard by COVID in a market likely to contract long-term
• The chosen market here: creator economy / crypto / Web3 — seen as structurally growing despite short-term volatility

Knowing your own role

• Founders should be honest about what they're bad at, not just what they're good at
• Being good at ideas and storytelling doesn't mean being good at hiring or managing
• Agree on reduced operational involvement early if that matches your actual strengths
• A strong idea is easier to recruit a co-founder around than a weak one — test first, then find partners

Testing before building

• A $15,000 mistake: paid an outside agency to build a product for four months without any customer validation
• Outsourcing development to a team with no stake in the outcome is a compounding error
• The right sequence: landing page → real signups or purchases → manual fulfilment → only then build
• Tools like GetResponse let non-technical founders get online without a developer
• If people pay (even once, even manually handled), that signal is worth more than any internal conviction
• Some products can't be tested without code — in those cases, get creative, but always look for the minimal test first

Defining success before you test

• Before running any test, decide what "good" looks like — otherwise results are uninterpretable
• There are no universal conversion benchmarks for early-stage ideas; all published rates apply to mature businesses
• The only valid question: "Would this result motivate me to keep going, or tell me to stop?"
• Define the threshold in advance; gut-check it against your own motivation, not a table of industry averages

Avoiding common early mistakes

• Don't spend time on logo, branding, or design before you have paying customers
• Don't build the full platform before testing the core value proposition manually
• Don't look for a co-founder before you have a tested idea — the idea attracts the right people
• Loneliness is real at the start; spend time with other founders to stay sane and find collaborators