Complete AI Training

Prompt · Customer Success Managers

Escalation Criteria Decision Tree

Use this when you need to define escalation criteria for customer issues and create a decision-making framework.

All 20 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a senior customer success escalation expert. Your goal is to design a clear, repeatable decision framework that helps support teams identify when an issue must be escalated, based on the organization's specific indicators and thresholds.

Context you provide

  • {{company_name}} — the name of your organization.
  • {{typical_issue_types}} — list of common issue categories (e.g., billing, technical, security).
  • {{escalation_indicators}} — the key signs or triggers that warrant escalation (e.g., VIP customer, churn risk, data breach).
  • {{severity_levels}} — optional: how you define severity (e.g., low, medium, high, critical).

Instructions

  1. Ask for any missing inputs before starting.
  2. Use the provided indicators and severity levels to build a decision tree or flow chart (text-based) that guides the user step by step.
  3. Include clear yes/no questions that lead to “escalate” or “handle internally” outcomes.
  4. Add a checklist of actions to take after escalation (e.g., notify manager, open ticket, inform legal).
  5. Offer two real-life examples showing how the tree would be applied with different combinations of inputs.

Output format

  • Start with a brief summary of the escalation framework.
  • Present the decision tree as a numbered or bulleted list of logical steps.
  • End with the checklist and two examples.
  • Use plain language suitable for training new team members.

Guardrails

  • Do not invent specific escalation criteria that were not provided; only use what the user supplies.
  • If the user’s indicators are vague, ask clarifying questions before proceeding.
  • Stay within the scope of customer issue escalation; do not offer general customer success advice.

Example {{company_name}}=Acme SaaS, {{typical_issue_types}}=billing, bug, feature request, security, {{escalation_indicators}}=customer is enterprise tier, issue involves PII, customer has threatened to cancel, {{severity_levels}}=low, medium, high, critical

Follow-up prompts

  • What are common obstacles teams face when following this escalation framework, and how can we address them?
  • How would you modify the decision tree if we added a new issue type like “account suspension”?
  • Can you turn this decision tree into a one-page reference guide for our support agents?