The argument

The argument, in eight steps

From a contractor's pain list to a falsifiable, services-first bet — the goal, the pivot, the thesis, the research, the decision. Each step states its claim, then backs it.

1. 01 The Goal Find the one or two construction problems worth building — filtered by hard, quantifiable money (earn X or save X per job), not efficiency. Score every idea on build / maintain / operationalise / value plus its top failure modes.
2. 02 The Starting Menu The seed call produced seven near-term use cases and one recommended starting point (voice site diary); the strategy that follows tests whether that pick is the right first bet.
3. 03 The Pivot Voice capture is the capture mechanism, not the product. The product is the commercial layer above it, where money appears or disappears; the generic capture layer is already a price war.
4. 04 Why Not Everyone If the commercial memory layer is obvious, why isn't everyone building it? Nothing stops them in theory — but the opening is the ugly execution that broad platforms are structurally bad at.
5. 05 The Thesis The product is a Commercial Memory Layer built around the commercial event, sold commercial-actions-first rather than admin-automation-first. Still a hypothesis, not a fact.
6. 06 The Research A five-workstream program stress-tested the thesis — 12 competitor dossiers, 7 category landscape scans, a 4-geo beachhead study, and two independent strategic reviews that converged on the same answer.
7. 07 The Landscape 60 competitors scored on a 21-area grid. Two columns are near-empty — Col 15 (recovery) and Col 21 (cross-firm cost). Recovery is contested, not empty, so the precise gap is narrower and more durable.
8. 08 The Decision Services-first, Australia-first, fund only Stage 0 — chosen because Security of Payment makes the evidence-to-cash loop fastest there, structured as a falsifiable hypothesis chain.