Measured, not asserted
Analect Proof is the measurement workflow in the Analect platform. It takes a document and its decomposition, asks the same questions of both against a model you choose, marks the answers against known facts, and totals the tokens each route consumed, so every claim about accuracy and cost is a number from your data rather than a promise.
Why proof comes first
Ninety-five percent of enterprise generative AI pilots show no measurable P&L impact, and the failures are integration-shaped: tools that never understood the data they were pointed at. A measured pilot on your own material is the antidote. Proof exists so the first conversation with your finance team is about your numbers, not ours.
Three steps to a number
Pick the data
A saved document from your library, the analysis you have open in Explorer, or a freshly uploaded file.
Pick the model
Any model from the catalog Forecast uses, not just the defaults.
Run the test
The engine asks the questions, marks the answers and totals the tokens. Results export as a report.
What we found on our own material
The zeroes are part of the report
Proof reports the zeroes as well as the wins. Whole-document workloads cost more on fragments, and the report will say so. That is the point: a result you can take to a CFO has to be a result that could have gone the other way.
Run the test on your documents.
A pilot runs the same questions against your documents and their fragments, marks the answers and totals the tokens, so the evidence is yours rather than ours.