Startup fundamentals
Organic startup ideas
Good ideas come from lived problems, not from a whiteboard labeled "Uber for X."
Bad startup ideas feel exciting in a coffee shop and die in a customer interview. Good ones often feel slightly embarrassing: they come from a pain you personally know. If you treat idea selection as "sit down and invent a unicorn," you will mostly invent solutions in search of a problem.
Paul Graham's argument in How to Get Startup Ideas still holds: live at the frontier of some domain and notice what is broken. You build something you yourself want.
The big idea
A problem book is a running list of irritations. When you or a friend complains twice about the same thing, write it down. Ideas that survive a month of irritation beat ideas that excite you for an afternoon.
Brainstorming sessions ("Uber for X") produce SISPs: solutions in search of a problem. They sound startup-like because they borrow a known success's shape. They usually fail because nobody has a hair-on-fire need, or the hard part is not the app.
The analogy of "just notice more" breaks if you never talk to anyone else. Lived pain is a starting point, not a market.
Walking through it
Twitter, Airbnb (air mattresses for conference attendees), Stripe (developers hated payments), Dropbox (USB drives and emailing files to yourself) — the pattern is lived frustration, not a whiteboard labeled "markets of the future."
Michael Seibel's order of operations:
- Pick a problem you care about.
- Brainstorm solutions with sharp friends (sometimes recruiting co-founders in the process).
- Ask why you are qualified.
- Ship a tiny MVP fast and do not fall in love with it.
Concept
A minimum viable product (MVP) is the smallest experiment that tests your riskiest assumption with real users. It is not "version 0.1 of the full vision." If the riskiest assumption is "will anyone pay," a landing page and a checkout button can be more valid than a half-built app. See MVP and scoping.
The MVP is a learning instrument, not a monument. Ideas morph. Airbnb started as air beds. Slack grew out of a failed game company. A good starting point with founder-market fit beats a mythical perfect pitch.
Link this to problems worth solving: intensity, frequency, and willingness to pay beat elegance.
Common mistakes
- Starting with the solution. You built a clever architecture or a pretty demo. Customers shrug. Kill the demo; keep the problem if it is real.
- Waiting for the perfect idea. Ideas change after contact with users. Waiting is how you never ship.
- Confusing boredom at work with a market. "Expense reports are annoying" is widely true and already owned by incumbents. Your insight has to be sharper than shared annoyance.
- Premature secrecy. Hiding the idea for fear of theft usually costs more in delayed learning than IP risk.
Check yourself
You have a polished "Uber for dog walking" deck and zero conversations with dog owners. What do you have?
A solution in search of a problem. Talk to owners before you write more slides.
When is a landing page a better MVP than a half-built app?
When the riskiest assumption is "will anyone pay," not "can we build it."
Key takeaways
- Prefer organic ideas from lived problems over brainstormed "Uber for X" shapes.
- Keep a problem book. Repeated irritation beats afternoon excitement.
- An MVP tests the riskiest assumption. It is not a tiny version of the vision.
- Ideas morph. A good starting point plus users beats a perfect pitch.
Further learning
How to Get Startup Ideas (Paul Graham)How to Get Startup Ideaspaulgraham.comMichael Seibel on problem books, MVP speed, and not falling in love with v0:
Next: schlep blindness.
Related
Startup ideas and why now
Where good ideas come from, which shapes to avoid, and how to judge an idea before you quit your job.
Finding problems worth solving
Pain intensity, frequency and willingness to pay — separating hair-on-fire problems from nice-to-haves.
Founder-market fit
An unfair reason to win this market — lived experience, distribution, or skill — often beats a fashionable idea.