Which acceptance checks should a reseller run on each sample drop?
Eight checks covering recomputation, calendars, policy tracing, status order, exception variety, identifiers and free-text scanning, with failures logged by identifier for the supplier.
A reseller can run these checks on each sample before agreeing to list it, and again on any later drop.
- Recompute every amended schedule and confirm the closing balance reaches zero within a stated rounding rule. - Confirm no due date falls on a non-business day unless the calendar rule explains the exception. - Trace twenty random amendments to their linked policy clause and check the decision matches. - Verify status transitions follow the permitted order, with no payments posted against cancelled schedules. - Count amendment types and outcomes to see whether refusals, escalations and reversals are present. - Check that identifiers are unique, stable across files and follow a documented synthetic style. - Scan free-text notes for details that look like real people, organisations or addresses. - Keep a record of failures with identifiers so the supplier can respond to specific rows.
A fixed set of checks run on every drop makes quality comparable across releases and gives the supplier specific rows to answer.
- Audience:
- Licensed resellers
- Content type:
- Checklist
- Buyer question:
- How to evaluate it
Content release: r-a220eb6e099b243b
Related questions
- How can a grader catch a quiet error in a premium instalment change?
- How do I check grounding and disclosure risk in generated data?
- How do I judge the quality of generated schedule data?
- How do state diffs, graders and hooks fit into a schedule evaluation?
- What success criteria should I set for a schedule agent evaluation?
- How does an evaluator catch rounding drift when an agent splits an instalment plan?