—.Proof — methodology behind the claims

The numbers, the way they were measured.
And what isn't measured yet.

Every specific number that appears on Forge — accuracy, speed, win-rate lift — comes here for its methodology. Sample size, comparison baseline, date range, stated limitations. The measurements that haven't happened yet are listed honestly with what's pending, instead of a confident number with no work behind it.

A claim with no methodology page behind it doesn't belong on the site. This index grows as the work does.

I
METHODOLOGY · MEASUREMENTS PENDING
Hyperion accuracy
The ±0.8% claim — methodology, sample plan, and what's pending.

Forge's published Hyperion accuracy claim — ±0.8% on commercial flat and low-slope roof geometry, ±2% on small-residential — and the methodology, comparison baselines, and sample plan that will turn the claim into a measured benchmark. Independent third-party verification is not yet complete; this page documents what's stated and what's pending so the rest of the site doesn't carry numbers that lack a published methodology behind them.

OPEN →
PENDING · NOT YET ON THE WALL
Charter operator outcomes — measured win-rate, tools replaced, payroll cycle time.
PENDING · NOT YET ON THE WALL
Contractor tool consolidation — methodology for the per-operator competitor stack replacement claims.
PENDING · NOT YET ON THE WALL
Hyperion field benchmarks — independent third-party measurement set, when complete.
THE COVENANT

A claim with no methodology page behind it doesn't belong on the site. The proof page is where the work shows.