Multi-season crop and yield history, built per parcel.
Every scored parcel carries a season-by-season crop record: what was grown, roughly how much came off it, and how that reads against regional patterns. Sources are stated on every reading - no single dataset carries the whole story.
Six seasons of yield, three data sources, one record.
A single record shown end-to-end. Real records vary by state and region - the shape below is representative, not universal.
Satellite-assisted
Growing-season vegetation indices at the parcel boundary, calibrated against ground-truthed regional yield curves.
Self-reported
Borrower or field-officer capture through a light mobile flow, stamped with the person who captured it and the time.
Agri third-party
District and block-level yield data from agriculture departments and licensed data providers, used as the calibration baseline.
Yield readings are estimates, and always labelled as such.
Where the platform is confident and where it is not - stated plainly, on the surface, not buried in a data sheet.
Yield is not directly measured off a satellite. What the platform reads is a growing-season signal and then a regional calibration against actual harvest data. This produces a stable estimate for the common cropping patterns of a region and a wider band for unusual ones.
Every reading on the crop-history strip is labelled with its data source and its confidence. First-season signals for a new crop are flagged as watch, not treated as established history.