home / portal / procurement_coverage_cells

procurement_coverage_cells: 27

The coverage matrix, one row per source x record kind x fiscal year, each carrying its state, its count, why it is in that state and the SQL that produced the count. Run the evidence_sql and you get n_records back.

This data as json

rowid source_key source_label record_kind record_kind_label fy_key state n_records prev_n_records why_code why_text evidence_kind evidence_sql evidence_citation evidence_lf_id probed_for_publication is_crawl_dependent drill_url computed_at run_id
27 gvibuy_purchase_orders GVIBuy purchase orders release Releases against a master blanket FY2025 PARTIAL 263 263 go-live-boundary 1 of the 263 documents in this cell were sent before the 1 June 2026 GVIBuy go-live, and the pre-go-live record is a fragment rather than a record. Counted the same way on both sides of that date: GVIBuy's purchase-order list holds 533 documents sent from 2022-01-19 to 2026-05-28 and 3,860 sent from 2026-06-01 to 2026-11-17, and of those, 43 before the go-live are transactional purchase orders against 3,492 after it - the remainder on each side being contract records, which are not purchases. DPP states why in its own words - 'No legacy POs will be migrated to GVIBuy; these will remain in Tyler Munis for their full lifecycle.' The year here is the year inside the identifier. It is a numbering year, not a spending year, so this cell counts RECORDS by when they were numbered and never when money moved. MEASURED SELECT COUNT(*) FROM procurement_po_classification WHERE record_kind = 'release' AND COALESCE(identifier_fy, 'no-fy-in-record') = 'FY2025'       0 /procurement/chain?source=gvibuy_purchase_orders&kind=release&fy=FY2025 2026-11-21T22:40:46Z 2026-11-21T22:36:15Z
Powered by Datasette · Queries took 2.021ms