Unit 09.02: The exceptions that are most of the volume

On real data the exceptions are not the edge. They are frequently most of the first pass.

Every exception needs a named destination

Triage sorts each record onto the normal path or to a specific destination.

The table below triages five invoices.

INV-1  normal path
INV-2  exception -> procurement: no purchase order
INV-3  exception -> finance: credit note, not an invoice
INV-4  exception -> a human: unsupported currency
INV-5  normal path

2/5 on the normal path (40%)

Three of five are exceptions, each with a named destination. The credit note is the instructive one: a negative amount is not a malformed invoice, it is a different document type that arrived in the same queue, and "handling" it would mean approving a payment of minus fifty.

An exception with nowhere to go is where an agent starts improvising, which is the failure this whole module exists to prevent.

The mistake this prevents

The mistake is discovering the exception rate after launch. Run the triage rules over a month of real historical records first - if 40% are exceptions, the automation's value is much smaller than the proposal assumed, and better to know before building.

Takeaway

Triage over real historical data before building, and give every exception a named destination. An exception with nowhere to go becomes improvisation.

Mark complete