ThinkWithPrompt by SynotticThinkWithPrompt

Set boundaries with an abusive or harassing customer

Calm, firm messages for each stage of abusive behavior, from a first warning to ending the conversation.

Customer SupportRTFStandardCommunicationSummarizeExplain & TeachFill-in Template

The prompt

Help me handle a customer whose behavior has crossed into abuse or harassment. What they've said or done (paste or summarize): [MESSAGES] Channel: [chat / email / phone / social] Their underlying issue: [WHAT THEY NEED RESOLVED] Our policy on abusive customers: [PASTE - or "none written"] What I've already said: [PREVIOUS REPLIES] First, classify the behavior: frustration (venting, swearing at the situation), abuse (insults, slurs, swearing at me), or threats (violence, doxxing, self-harm). Explain the difference briefly. Then give me: 1. A message that acknowledges their issue and sets a clear boundary, naming the behavior, not the person. 2. A final-warning message stating what will happen if it continues. 3. A message ending the conversation, with how they can get help through another route. 4. A factual internal incident note: quotes, times, actions taken. 5. When to involve my lead immediately. Rules: - Stay neutral and short. Don't match their tone, argue, or explain my feelings. - Keep working on their real issue as long as it's safe to. - Threats of violence or self-harm go to my lead and our safety process right away; if someone is in immediate danger, contact local emergency services. - I don't have to tolerate abuse to be good at my job.

Highlighted [brackets] are blanks to replace with your own details.

Fill in the blanks and use this promptMore Customer Support prompts

What to fill in

  • MESSAGES
  • chat / email / phone / social
  • WHAT THEY NEED RESOLVED
  • PASTE - or "none written"
  • PREVIOUS REPLIES

How this prompt is built

This prompt follows the RTF structure (Role, Task, Format). RTF tells the model who to be, what to do and what shape the answer should take. It is the quickest structure that still gives noticeably better results than a bare request.

Related prompts

Find more prompts like this