If anyone building a product outside the US thinks they’ll eventually expand here — you’re not going to make it.
I built an AI research analyst for investors, at my last company. We tried validating it in Korea first, then going for the US. It ultimately failed. Now I’m building Char in SF — and it’s been a completely different experience.
Product development is iteration. You need people who are eager to try your half-broken thing and actually give you feedback. Koreans are picky — they want the polished version. People on Reddit were almost identical. Neither group wants to touch your v1.
SF is different. People here valorize the builder spirit. They want to try what you’re making because it’s new, not in spite of it. It’s a sandboxed playground for the world’s most cracked founders — and the early adopters willing to bet on them.
In the earliest stage of a startup, you’re building relationships with your customers. That takes time. Once you earn their trust, there’s inertia. If you aren’t a psychopath, you naturally want to serve them better. You get locked in. So where you start those relationships matters more than anything.
The hard truth: you have to start in the States. Especially SF, if you’re doing software. You need to talk to SF folks.
SF is the left side of this curve. Innovators and early adopters live here. If you’re not starting where they are, you’re skipping the only people who’ll give your half-baked v1 a real shot.
So stop framing it as “global expansion to the US”. Start from SF. Expand outward.
But — if you can’t eventually leave SF and cross the chasm, you’re fucked too.
The exception: if you already have an audience that’s almost identical to your end users, you can start wherever they are. Most people don’t have that. So start from SF.
April 12, 2026