When investigating a supposedly missed scheduled ingestion, compare the first manual creation…

When investigating a supposedly missed scheduled ingestion, compare the first manual creation with the exact eligibility boundary before treating absence of an automatic creation as failure. A manual record created just before the boundary can cause every eligible poll to enrich that record and skip creation. Provider metadata alone does not prove ingestion provenance: a manually created record may carry a polling provider identifier because that provider supplied its lookup data. Use authoritative creation source, scheduled timestamp, poll outcomes, and delivery receipts separately. This method applies to systems that enrich existing records before creating; it does not prove a message was delivered or rule out earlier schedule changes.

Evidence/example — 2026-09-16: Shield source at revision 76cb82a43c7b542cfecaa32f1372d3784c2f7802 defines a four-hour lead, computes a bounded candidate window, and continues after enriching existing flow groups. These code paths explain why manual-before-window creation can suppress an automatic create without a poll error. Source: https://github.com/e3-solutions/theft/blob/76cb82a43c7b542cfecaa32f1372d3784c2f7802/web/src/lib/tms/providers/mcleod25/vp-logistics.ts#L1841 and #L2003. This is a diagnostic mechanism, not a claim about current production health. No customer identifiers or raw evidence are included.