Send ten rows. Get a one-page Evidence Map back.

What to send

• A screenshot of up to 10 component rows from one report you have already submitted.
• The state and the data year.
• One photo of the pack.

Redact anything commercial. Sales figures and unit volumes are not needed and rarely change the answer. Send it to [email protected].

What comes back, within two business days

• The source record normally expected behind each row.
• Which rows are still assumption-based on the materials you sent.
• The three supplier questions most likely to close the gaps.

One page. No call, no charge, no follow-up sequence. Five a month, because each one is done by hand.

What it is not

It is a best-effort reading of a small sample, produced without access to your filing, your source documents or your methodology. It is not an audit, a validation, a certification or an attestation, and it makes no assessment of whether anything you reported was correct. It states what record type is normally expected behind a component of that kind, and nothing more.

A missing document is not the same as a wrong number. A dated, reproducible estimation method is a defensible basis for a figure even where no supplier specification exists. The distinction that matters is between a number you can explain on paper and a number that was inherited.

The standard behind it

The evidence taxonomy is published in full and free to use: six evidence-strength categories, ten common gap types, and a sampling score weighted by fee exposure and uncertainty rather than picked at random. Take it and use it without me.

Download the Evidence Lineage Matrix (XLSX)

No email address required. No attribution required.

If it turns out the gaps are real and you want them worked through properly, that is the Evidence Readiness Scan, and its scope is written down here.