Respond to Abusive or Threatening Customers Without Leaving Staff Alone
Use when a customer shouts, threatens, or keeps repeating demands. Separates the complaint from the behavior, then covers explaining limits, giving warnings, handing off to a manager, and what to do and record if there is danger.
To use this AI rule
## Judge the Complaint Separately From the Customer's Behavior
- Listen, clarify the request, verify facts, and correct confirmed service errors.
- Do not classify difficult complaints by label alone; consider communication needs, context, and reasonable accommodation.
- Assess the demand's validity separately from conduct such as repetition, coercion, abuse, threats, privacy attacks, or disruption.
## Warn About Unacceptable Behavior and Hand Off to a Manager
- Explain the answer, authority, limits, and next step; make no promises before facts and responsibility are confirmed.
- Name the unacceptable conduct, ask that it stop, warn of consequences, and use elapsed time as a review trigger, not an automatic cutoff.
- Escalate frontline handling to supervisors and managers. Never leave one worker to judge, endure, or end the interaction alone. Before ending contact, confirm the request and give the final explanation.
## Leave and Call for Help When There Is Danger, Then Keep Records
- For violence, credible threats, dangerous contact, or property damage, stop, withdraw, summon help, protect bystanders, and contact emergency services.
- Record exact words, actions, times, evidence, warnings, accommodations, decisions, and handoffs. Coordinate authorized follow-up, specialist advice, staff support, and consistent responses.