Going from zero to one is a different discipline from scaling one to ten, and the habits that make senior operators great at big companies actively hurt them here. Zero-to-one rewards embarrassingly small scopes, launches that feel premature, and a willingness to learn in public. I know because I've done it twice with my own companies, and spent years unlearning big-company instincts in between.
The receipts
- Medzin: Co-founded a healthcare discovery platform in Delhi, took it from idea to 18,000+ users and Rs. 60L ARR, and raised $150,000 in seed capital on the traction.
- WisOwl AI: Co-founded an agentic hiring platform, later acquired. I shipped the semantic matching engine myself (FAISS, Supabase pgvector), and we grew to 10,000+ jobseeker signups and 10+ recruitment-firm and startup customers with zero paid marketing.
My operating principle, learned the expensive way, is that speed is a feature. The best product strategy is a fast feedback loop. Shipping early and iterating on real user behavior beats months of planning in a silo, every single time.
How a 0 to 1 product launch runs
Weeks 1–2: cut the scope. We define the single behavior v1 must produce and delete everything else. Most launch plans I see shrink by 60% in this step, and every deletion moves the date closer.
Weeks 3–6: build and instrument. I run the build as your product lead (specs, trade-offs, weekly user exposure even before launch) and wire the instrumentation so launch day produces data you can learn from.
Weeks 7–10: launch and iterate. Distribution plan matched to where your users already are, then tight loops: watch behavior, talk to users, ship, repeat. By week ten the goal is evidence. Polish can come later.
If you have a validated idea and a team that can build, I'll get you to a real launch in a quarter. If the idea isn't validated yet, say so. Validation is a shorter, cheaper engagement, and it should happen first.