Ask any real estate developer how long it takes to know if a site is viable, and the honest answer is still months — usually four to twelve of them, and usually across four separate firms: a broker, an architect, an environmental consultant, and a bank underwriter, each working from their own version of the data.
The actual bottleneck isn't analysis. It's handoffs.
None of the individual steps in a feasibility study are slow in isolation. Zoning lookup, massing studies, cost estimation, underwriting — each of these can run in hours with the right tooling. What actually costs the months is the handoff between them: a broker's site data gets re-entered by an architect, whose massing options get re-keyed by a cost estimator, whose numbers get re-verified by an underwriter who was never in the room for any of the earlier decisions.
Every handoff is a place where context gets lost, assumptions drift, and someone has to re-derive facts the previous firm already established. Multiply that across four handoffs and a trillion-dollar industry ends up running on PDFs, spreadsheets, and consultant loops — not because any single step is hard, but because nothing in the chain talks to the next link natively.
What changes when it's one model, not four
The fix isn't making any individual step faster. It's collapsing the chain so the same underlying site model flows from land intelligence through masterplanning through cost estimation through underwriting — without a human re-entering data at each boundary.
That's the actual bet behind Keystone: land intelligence, generative planning, and feasibility/underwriting running on one connected model, with India (RERA/NBC/CPWD) and USA (IBC + all-50-state FAR data) regulatory logic built in natively rather than bolted on per-market. A masterplan option isn't just a drawing — it's already checked against the zoning and cost constraints that would otherwise take a separate consultant to verify.
What this doesn't mean
It doesn't mean fewer questions get asked. If anything, a live model invites more scrutiny, not less — you can change a parameter and immediately see what breaks, instead of waiting three weeks for a revised PDF. And it doesn't replace the judgment calls a developer or lender still has to make. It removes the mechanical cost of re-deriving the same facts four times before anyone gets to make those calls.
We're early. Keystone is pre-revenue, live on real sites, with pilot conversations in progress across India and the USA. But the direction is clear: the industry doesn't need a faster version of the old process. It needs one that doesn't require four handoffs to answer one question — is this site worth building on.