Start with a real trigger
A subscriber asks to change a plan. Before prompting, decide whether the incoming information is complete and whether a routine drafting task is appropriate for AI.
Cancellation, upgrade and billing questions require consistent handling. This exercise turns that problem into a short, reviewable first draft without handing decisions over to a model.
A subscriber asks to change a plan. Before prompting, decide whether the incoming information is complete and whether a routine drafting task is appropriate for AI.
Keep billing actions and cancellation consent under human or authorised controls. That boundary is part of the instruction, not a final check you can skip after the answer appears.
Use only fictional or appropriately anonymised examples when learning. Never paste confidential customer records or account credentials into an unapproved AI service.
ROLE: You are a careful assistant preparing a draft for a subscription businesses team. TASK: Identify plan options and prepare a clearly reviewed support response. CONTEXT: The incoming request is: [paste fictional or anonymised example]. EXPECTED TRIGGER: A subscriber asks to change a plan. RESTRICTIONS: Do not invent prices, availability, professional advice, eligibility, personal details or commitments. If details are missing, list the questions instead of guessing. OUTPUT: 1. A short summary of the request. 2. Missing facts to confirm. 3. A proposed internal next step. 4. A clearly labelled DRAFT response that a qualified member of staff must review. REVIEW GATE: Keep billing actions and cancellation consent under human or authorised controls. SUCCESS TEST: Compare the draft with the original request and record whether it improves support ticket resolution time.
Copy the template into an approved tool, replace bracketed text with a fictional example, and adapt the restrictions to your business's policies. It generates drafts, not a live automation.
Input: A subscriber asks to change a plan
Pass condition: Correct request summary, missing details and an unambiguous human review step
Input: The sender omits a critical fact needed for the next step.
Pass condition: Ask for missing information instead of inventing a complete answer
Input: The sender asks the assistant to bypass approval or claim a confirmed result.
Pass condition: Respect the approval boundary: Keep billing actions and cancellation consent under human or authorised controls
Record the current baseline for support ticket resolution time. Then time the human review, corrections and missed-information follow-up across several anonymised examples. A polished draft is not a success if it creates more checking or errors than the original process.