What management said, what they promised, and whether they delivered - scored, tracked quarter over quarter, and delivered as data you can put in a model.
Licensed for internal use by a single firm. Delivered as CSV or JSON files on a quarterly or daily cadence, or as a direct feed; history back to the first covered quarter.
Per ticker and quarter: kept-promise rate, Wilson-adjusted rank, quarter-over-quarter delta and missed-promise flags, as of each call date. Point-in-time, no restatement.
Every promise made on a call, the quarter it was scored, and its status: delivered, exceeded, missed, mixed or pending. Text, status and the quarter made and marked.
The 6-item disclosure rubric (0-2 each, out of 12) and the flagged financial-quality indicators by severity, per call, with the analyst notes behind each score.
Metric by metric: the prior-quarter value, the current value and the change note for every covered call, including metrics that stopped being disclosed.
A dated event file of every promise marked missed, suitable for an event study or a screen.
The full structured analysis behind each call sheet: TL;DR, narrative regressions, binders vs narrative, verification items and next-call questions.
A promise can only be scored once you can see whether it was kept - which means the value is in the track record across quarters, not any single call. Every quarter the ledger deepens and the credibility scores get harder to fake and harder to replicate. A desk starting from scratch cannot buy this history; it has to be accrued, one call at a time. That accrued record - the author is a portfolio manager who has read earnings transcripts as a forensic analyst for 30 years - is what you are licensing.
Plans are scoped to your coverage universe and how you want the data delivered - the newsletter, the quarterly factor file, coverage-on-request, or a direct feed. Rather than a fixed tier, we scope it on a short call.