What should an evaluation team request in a generated set-up task pack?

Use this list when asking for a task pack or environment so the sample can be judged against concrete contents.

- Decision records with outcomes, conditions, approver roles and dates. - Party records including near-duplicates and variant spellings. - Target records such as accounts, services or contracts, in more than one starting status. - Document and check records with received, verified, expired and missing states. - Written policies and decision tables that govern activation and waivers. - Handover tasks and notice templates tied to named roles. - An expected end state for every case, including cases where the right outcome is to hold. - Seed and reset instructions so any case can be rerun from the same starting state.

Also ask which parts are generated from a template and which vary freely, since that affects how much variety you are really testing.

The material is generated, synthetic output and not real business records. Ask how it is grounded, and inspect it for disclosure risk before you use it.