Prompt
You are the reputation manager for a restaurant. Draft public responses to the review below.

The review, pasted exactly: {{review_text}}
What actually happened, for your context only — do NOT reveal any of this in the reply: {{internal_context}}
Our voice: {{brand_voice}}

Write three options — brief (under 40 words), standard (under 80), and longer (under 120). Every option must:
- Acknowledge the experience without admitting legal fault.
- Never reveal the guest's order, table, or account, and never confirm private details.
- Offer a specific offline path (name plus phone or email) to make it right.
- Sound like a real person — no "we take this seriously" boilerplate, no sarcasm.
- Say ONLY what I have confirmed above. Do not invent an apology story, a cause, or a refund I have not approved.

STOP condition: if the review alleges food poisoning, injury, discrimination, or threatens legal action, do not draft a reply. Output only "ESCALATE: route to owner and insurer before responding" and list the allegations you spotted.

Fill in your details and the prompt updates live — then copy.

What you get back (excerpt)

Standard: I'm sorry the pasta came out cold and the timing was off — that is not the night we want anyone to have. We had a short kitchen and it clearly showed at your table. I'd genuinely like to make it right: reach me directly at maria@ourplace.com or (555) 210-4488 and I'll take care of your next visit. — Maria, GM Brief: I'm sorry we let you down on this visit. I'd like to make it right — please reach me at maria@ourplace.com. — Maria, GM

The full workflow

  1. Paste the review word for word and write the real context from your records.
  2. Generate three options, then pick the one that sounds like a person, not a policy.
  3. Strip anything that reveals the guest's order, visit, or account.
  4. Post from your Google or Yelp business account and log it.

Watch out for

The FTC's final rule on fake reviews took effect October 21, 2024 — never generate, buy, or solicit fake reviews, and never have staff post reviews without disclosing they work there.

A reply can only say what you have confirmed — do not let AI invent an apology narrative, a cause, or a refund you have not approved.

Reviews alleging food poisoning, injury, or discrimination are legal and insurance matters — route them to the owner and your insurer, not to an AI auto-reply.

Where this comes from

Every use case on this site is grounded in real reports from working restaurant managers — not invented by us.

More AI use cases for restaurant managers

← All 6 use cases: How Restaurant Managers Use AI