Policy-Safe Support Macro Generator (Email / Chat / Phone)

One-shot prompt template that produces channel-ready, compliant agent reply macros, escalation triggers, QA checks, and KB-ready content for SaaS/ecommerce support teams.

What You'll Get

A single one-shot prompt template (placeholders) that outputs Email, Live Chat, and Phone macros.

Two macro variants per channel (Empathetic & Firm) with handling-time estimates.

Required data slots, 3-6 prioritized diagnosis questions, and ready-to-send response drafts using {{customer_name}} and {{ticket_id}} tokens.

Optional quick-insert snippets (apology, escalation handoff, compensation phrasing) and a single-sentence Next Steps line per macro.

Explicit escalation triggers with severity labels, exact role/team to contact, and time-to-escalate SLAs.

A 5-item QA checklist and a clear Facts vs Assumptions block for each channel output.

Markdown-formatted output ready to paste into agent tooling, KBs, or playbooks.

How to Use

1Replace placeholders ({{product_service}}, {{customer_tier}}, {{brand_voice}}, {{policy_rules}}, {{issue_type}}, {{customer_sentiment}}, {{resolution_options}}) before running the prompt.

2Ensure policy_rules and resolution_options are explicit. If any required input is missing, the prompt will ask a clarifying questionâ€”provide it before continuing.

3Run the one-shot prompt in ChatGPT, then validate generated macros against internal policy and a small pilot sample of live tickets.

4Copy the Markdown sections into your agent toolbox, update role/team names and SLA windows to match your org, and import approved snippets into canned response systems.

5Periodically re-run with updated incident/context and use the Facts vs Assumptions block to record live changes.

Mandatory: supply policy rules and resolution options

Before generating macros, provide explicit {{policy_rules}} (refund limits, PII rules, escalation authority) and concrete {{resolution_options}}. Without these the output will either ask for clarifications or make conservative assumptionsâ€”always validate the macros against legal/compliance and pilot with a small agent cohort.

Works Best With

Best used by support leads to generate channel-ready macros and operational SOPs

Provide complete input values for: issue_type, customer_sentiment, resolution_options, and policy_rules to avoid clarification prompts.

Include recent incident context and any relevant customer history for higher-fidelity macros.

Use for both agent-facing macros and copy-paste-ready customer-facing replies (email/chat/phone).

Ideal for teams that need consistent escalations: include role/team names and SLA targets when running prompt.

Use brand_voice to control tone variants (e.g., "concise & empathetic", "friendly & helpful").

Prompt Template


## Goal
- Produce policy-safe, channel-specific agent reply macros for {{product_service}} and {{customer_tier}} that support email, chat, and phone interactions. Each macro set must include diagnosis questions, response drafts (empathetic and firm variants), required data slots, optional snippets, escalation triggers, and a short QA checklist so outputs are distribution-ready.

## Inputs
- Context: product details, recent incidents, customer history.
- Stakeholders: support agents, escalation engineers, legal/compliance.
- Constraints: regulatory or internal policy rules, permitted refunds/credits, SLAs.
- Desired tone: defined by {{brand_voice}} (e.g., â€œconcise & empatheticâ€).
- Required format: channel-specific macros with templates and clear data slots.
- If any of {{issue_type}}, {{customer_sentiment}}, {{resolution_options}}, or {{policy_rules}} are missing, ask a clarifying question before generating macros.

## Prompt
You are a customer support macro generator for the product/service described below. Produce channel-ready, policy-safe reply macros and associated operational materials suitable for copy/paste into agent tooling.

Inputs (use these values exactly as provided):
- Product/service: {{product_service}}
- Customer tier: {{customer_tier}}
- Brand voice: {{brand_voice}}
- Policy rules: {{policy_rules}}
- Issue type (single phrase): {{issue_type}}
- Customer sentiment (choose one): {{customer_sentiment}} â€” e.g., neutral, frustrated, angry, confused
- Resolution options (bullet list): {{resolution_options}}

Task requirements:
1. For each channel (Email, Live Chat, Phone) produce two reply macro variants: Empathetic and Firm. Each macro must include:
   - A short descriptive title and estimated handling time.
   - Required data slots (field names agents must collect).
   - Diagnosis questions (3-6 concise, prioritized).
   - Full response draft ready for the channel (use {{customer_name}} and {{ticket_id}} tokens where applicable).
   - Optional short snippets for quick insertion (apology, compensation offer, escalation handoff).
   - A clear â€œNext stepsâ€ line the agent should send to the customer.
2. Provide explicit escalation triggers with severity labels and exact next-person-to-contact (role or team), including SLA time-to-escalate.
3. Provide a short QA checklist (5 items) the agent or reviewer must confirm before sending any macro-based reply, including compliance with {{policy_rules}}.
4. Mark any assumptions you make; separate facts (from inputs) vs assumptions (derived).
5. Output must be formatted with clear Markdown headings and bullet lists, ready for agents and support ops to distribute.
6. Enforce the Support output format: for each macro include â€” Diagnosis questions â†’ Response draft â†’ Next steps.

Generate the full artifact now. Keep language concise, actionable, and strictly policy-safe.

## Output Format
Support: Diagnosis questions â†’ Response draft â†’ Next steps

For every channel macro include:
- Title Â· estimated handling time
- Required data slots (list)
- Diagnosis questions (priority-ordered)
- Response draft (Empathetic / Firm)
- Optional snippets (short bullets)
- Next steps (single-sentence customer message)
- Escalation triggers (condition â†’ role/team â†’ SLA)
- QA checklist (5 checks)
- Facts vs Assumptions (short two-line split)

Produce separate sections labeled Email, Live Chat, and Phone. Ensure each section uses the Support format above.

## Guardrails
- Separate Facts vs Assumptions clearly at the end of each channel section.
- Never promise refunds, credits, legal outcomes, or SLA exceptions unless explicitly allowed in {{policy_rules}}. If a resolution option conflicts with {{policy_rules}}, flag it and suggest compliant alternatives.
- Do not include customer PII except allowed tokens like {{customer_name}} and {{ticket_id}}.
- Escalation triggers must be actionable (role/team names, not individual persons) and include time-to-escalate.
- If required input fields are missing, request them before generating macros.

## Examples
- Mini excerpt (Email, Empathetic)
  - Title: â€œRefund inquiry â€” Empatheticâ€ Â· 10 min
  - Required data slots: {{customer_name}}, {{ticket_id}}, purchase_date, payment_method, order_number
  - Diagnosis questions: 1) Can you confirm the order number? 2) When did you purchase? 3) Describe the issue in one sentence.
  - Response draft: â€œHi {{customer_name}}, Iâ€™m sorry youâ€™re experiencing this. I can help investigateâ€”could you confirm your order number and the payment method used? Once I have those, Iâ€™ll check eligibility under our policies and follow up within 48 hours.â€
  - Next steps: â€œPlease reply with the order number and payment method and Iâ€™ll investigate right away.â€
  - Escalation trigger: if customer reports unauthorized charge â†’ Escalate to Payments Investigations within 2 hours.
  - QA checklist: 1) Policy check for refund eligibility 2) No promises outside {{policy_rules}} 3) Tone matches {{brand_voice}} 4) Include required slots 5) Ticket ID present
  - Facts vs Assumptions: Facts â€” product/service = {{product_service}}; Assumptions â€” customer provided no order number

Variables

Usage Notes

Not for

â€¢ Legal advice, regulatory interpretations, or binding contractual language.

â€¢ Medical, psychiatric, or emergency crisis responses.

â€¢ Financial-product regulatory decisions (e.g., lending, insurance claims) unless vetted by compliance.

Do not include

â€¢ Customer PII beyond allowed tokens (do not embed actual emails, SSNs, card numbers).

â€¢ Unapproved promises (refunds/credits/legal outcomes) unless explicitly permitted in provided {{policy_rules}}.

â€¢ Individual employee contact namesâ€”use role/team names only for escalations.

â€¢ Internal system prompts, model-spec details, or hidden instructions.

Policy-Safe Support Macro Generator (Email / Chat / Phone)

Create policy-safe, channel-specific reply macros and operational artifacts (Empathetic + Firm variants), including data slots, diagnosis trees, escalation triggers, and a compliance QA checklist.

Turn this into a working asset in minutes.

Import into Workspace to auto-fill variables, adapt output style, and iterate with OctiAI's prompt editor.

Auto-fill & personalize from your inputs

AI-assisted edits: tone, structure, constraints

Attach context so outputs stay consistent

Iterate and save versions

What you'll get

Output format

Markdown

Best used when

When working on template or detailed.

Avoid when

Legal advice, regulatory interpretations, or binding contractual language.

Made for

What to fill in

Use the guidance below to complete each placeholder with the right level of detail.

Product Service