Training and certification records joined to the scorecard of the person who took them, before and after. A course either moves a number or it is a line item.
Reference scale, grade Derived. Not a forecast for your business. This figure describes what a decision of this shape has been worth at this scale of operation, drawn from the evidence named in section 02. No payback, return or year-one figure appears in this brief — no fee has been quoted, so those numbers cannot exist yet. They are computed against a signed quote and stated with the fee beside them.
Training gets approved because it is reasonable and the budget holds it. Completion is tracked. Whether the trained person then sold, built or closed differently is not.
The training budget gets defended on completion rates, which measure attendance.
Training records carry a date and a person. Rep scorecards carry the same person and a monthly result. Joined, they say which courses moved the metric they were bought to move, which moved nothing, and which had an effect that decayed inside two quarters.
The decayed ones are the useful finding — they are working courses on the wrong cadence, not wasted spend.
The rest of this model — the data sets it runs on, the cause split, the working behind the reference figure and a sample output — goes out by email. Two fields, nothing else.