The problem
If you own a lot and wonder "could a pool fit back there? what would it run me?" — the honest answer today is: book a site visit, wait for two or three contractors, and compare quotes that aren't scoped the same way. That's weeks of latency for what is fundamentally a geometry-and-pricing question.
Approach
The pipeline is deliberately boring — no black boxes, just authoritative public data:
ADDRESS ─▶ GEOCODE ─▶ COUNTY PARCEL ─▶ LOT GEOMETRY ─▶ FEASIBILITY ─▶ COST BY TRADE
(Census) (assessor GIS, (lot, yards, (pool / ADU / (excavation,
exact lot lines) setbacks) garage, per concrete,
local rules) electrical…)
- Locate — the address is geocoded through the US Census, then matched to the county assessor's authoritative parcel record. The exact lot polygon is drawn in red over current Esri satellite imagery, stamped with the photo's capture date so you know how fresh it is. When a lot line can't be confirmed, it says so rather than guessing.
- Measure — lot size, building footprint, yard space, and setbacks are computed from the real parcel geometry projected to meters — measured, not eyeballed.
- Judge — candidate improvements (pool, garage, ADU, addition, shed, sport court) are scored against the buildable footprint and setback rules, returning a plain feasible / possible / no verdict with the constraining factor called out.
- Price — each project is broken down by trade — excavation, concrete, framing, electrical, plumbing — against regional cost baselines that auto-escalate each year, so the estimate reads like a contractor's bid instead of a single mystery number.
What it outputs
A one-page report per address: the outlined parcel on live imagery, a space breakdown, a feasibility verdict with clearance checks and risk flags, and a materials-and-labor cost range tuned to your region and finish level — plus a spot to log real contractor bids against the model. The goal is that a homeowner walks into a contractor conversation already knowing the shape of the answer.