What should we ask for in a sample or pilot of generated booking material?
A checklist of items to request in a booking sample: representative cases, answer keys, policy documents, fault injection, a reset path and a defined review process.
Ask for items that let your team run and grade the material, not only read it.
- **Representative cases:** bookable, blocked and ambiguous cases for one chosen product. - **Answer keys:** expected end states, acceptable alternative paths and required handoff facts. - **Policy documents:** authority rules, pricing tiers and conditions lists tied to rule IDs. - **Fault injection:** timeouts, rejected calls and conflicting documents, switchable per task. - **Reset and snapshot path:** a command that restores each task's starting state and exports before and after state. - **Schema and manifest:** field definitions, key relationships and the file layout. - **Review notes:** a way to send back disputed keys or implausible cases and see them addressed. - **Licence summary:** the permitted uses, stated plainly, before you load the material into any pipeline.
Also name the question the pilot must answer, such as whether the agent handles amended approvals, so that the sample can be shaped around it.
Requesting these items turns a sample into something you can run, grade and review rather than only read.
- Audience:
- Agent developers
- Content type:
- Checklist
- Buyer question:
- Scope and getting started
Content release: r-a220eb6e099b243b
Related questions
- What does a first pilot look like for an agent team in vendor finance?
- What limits apply, and will we receive any real customer records?
- How could a pilot be delivered, versioned and wired into our CI?
- How should a team scope and license a generated booking pack?
- How should a sample be delivered, versioned and load-checked in my pipeline?
- What does a first sample review look like for an environment team?