Why 90 days, and why speed is a moat in India
In the Indian market, the gap between a validated idea and a live product is where most ventures die — burning runway on research, over-engineering, and indecision. Sitio Labs runs a deliberate 90-day clock because in a market this fast-moving, a working product earning real user feedback beats a perfect product launched six months late. Speed is not recklessness; it is a forcing function that prevents teams from building features no Indore shopkeeper or Chennai clinic actually wants. The 90 days are structured into three thirty-day phases, each with a hard, demonstrable outcome.
Days 1–30: validation and the riskiest-assumption test
The first month is not for building — it is for killing bad assumptions before they cost money. We spend weeks one and two interviewing 30 to 50 real prospective users in their language, whether that is Marathi MSME owners or Tamil-speaking clinic staff, and identifying the single riskiest assumption the venture depends on. Weeks three and four go to a throwaway prototype — often a Figma flow or a WhatsApp-based concierge MVP — to test willingness to pay before any production code exists. If the riskiest assumption fails here, we pivot or stop, having spent thirty days instead of three hundred.
Days 31–60: building the AI-native core, not the whole product
The second month builds only the irreducible AI core that makes the product work — the loan-underwriting model, the vernacular voice agent, the document-extraction pipeline — and nothing more. We deliberately resist the urge to build settings pages, admin dashboards, and edge cases, because the goal is a product that does one valuable thing exceptionally well. Alongside the model we stand up an evaluation harness from day one, since an AI product you cannot measure is one you cannot improve. By day 60 there is a working, instrumented product that a small set of design partners can use for real.
Days 61–90: design partners, iteration, and a real launch
The final month puts the product in front of 10 to 20 paying or committed design partners across the target Indian segment and iterates on what breaks in the real world — latency on a 4G connection in Nagpur, a dialect the voice model mishears, a workflow that does not match how a kirana store actually operates. We ship updates weekly and watch the evaluation metrics and retention signals rather than vanity downloads. By day 90 the venture has a live product, paying users, and the data flywheel turning. The output is not a demo; it is a company in market with evidence it should keep existing.
What makes 90 days achievable — and what it is not
The playbook works because the studio brings pre-built infrastructure — auth, payments via UPI and Razorpay, model evaluation tooling, and deployment pipelines — so the team spends its 90 days on the unique 20% rather than reinventing the commodity 80%. It is not a promise of profitability in 90 days, nor a substitute for the years of iteration that follow; it is a disciplined sprint to a validated, instrumented product in market. For founders, the value is brutal clarity: in three months you know whether you have something real. That clarity, delivered fast, is the single most valuable thing a studio can give an early venture.