Build something.

I don’t have an idea yet

That’s a fine place to start.

Don’t wait for a perfect startup idea. A strong first business can be a service, product, resale concept, local business, digital product, creative offer, or small software tool. Aim for an idea you can test, not one you can admire.

Start with these questions

• What are you unusually good at, or willing to learn?
• What do people already ask you for help with?
• What group of people do you understand well?
• What recurring problem annoys you or the people around you?
• What could you sell or test with less than $100?
• What could you launch in one weekend?

Then generate ideas fast

Crazy 8s uses eight boxes and eight minutes. The time limit pushes you past your first obvious ideas. SCAMPER takes something that already exists and asks what you could substitute, combine, adapt, modify, put to another use, eliminate, or reverse. Pick one idea to test.

I have an idea

Before you build anything big, find out whether a real customer cares.

• Talk to people who actually match your target customer.
• Ask about what they do now and what frustrates them, not just “Would you buy this?”
• Look for evidence: existing spending, workarounds, repeated frustration, time lost, or people actively searching for a solution.
• Test your riskiest assumption first.

When people want it but you don’t know what to build, define the smallest useful version: the main user flow, the features, and what you will leave out for now.

I want to start a business

Validate, price, test, and move.

The Member Startup Builder Guide helps you work out your stage, your biggest uncertainty, and one action you can take this week. It walks through validation, pricing, your first customer, and the economics of each sale.

Questions worth asking yourself

• Why does somebody care about this?
• Who is the first real person I would sell to?
• How would I reach five of those people?
• What does one sale cost me to deliver?
• What can I finish in the next seven days?

I want to build software

Personal software projects belong here, too.

AI coding agents can help you build software at real scale. The skill is telling the AI exactly what to build. The path:

1. Write a PRD (product requirements document) that says what to build, for whom, and what counts as done.
2. Ask your coding agent for an implementation plan before it writes code.
3. Let the agent build from the approved plan.
4. Test what it made against your acceptance criteria.
5. Save it to GitHub.
6. Launch, learn, and iterate.

I’m already building

Find the one thing holding you back right now.

Name your current bottleneck, then run one focused, measurable experiment on it. Common bottlenecks: customer acquisition, pricing, product, validation, operations, retention, capacity, or technical execution.

I want feedback or collaborators

Come to a meeting and bring what you have: an idea, an early draft, a running business. You’ll meet other builders, and you can ask an officer for a consultation.

You can also reach us on Instagram.