← Back to Horizon Flow resources
AI search trust signals · review operations
AI Search Review Response Policy Checklist for Small Businesses
Public review replies are becoming source material for buyers and AI answer engines. Use this checklist before letting an owner, VA, or AI assistant reply to Google, Yelp, Facebook, marketplace, or industry-directory reviews.
When to use this worksheet
- You reply to reviews from more than one inbox, location, or team member.
- You want review replies to sound helpful without admitting facts that are not verified.
- You use AI to draft replies and need a human-review stop rule before posting.
- You want AI search tools to see consistent service standards, escalation paths, and contact facts.
Copy/paste review-response policy worksheet
Review source: Google / Yelp / Facebook / marketplace / industry directory / other
Reviewer type: verified customer / unknown / vendor / employee / spam suspected
Review sentiment: positive / neutral / negative / safety-sensitive / refund-sensitive / legal-sensitive
Known facts we can safely confirm: [appointment date, service type, ticket number, store/location, staff member if public and appropriate]
Facts we must not invent: customer identity, diagnoses, warranty coverage, refund approval, pricing exceptions, timeline commitments, private account details
Escalation owner: [owner/manager name or role]
Private follow-up path: [support email, phone, form, ticket link]
Public reply goal: thank / clarify next step / invite offline resolution / correct public info without arguing
NEEDS OWNER REVIEW if: refund, warranty, safety, discrimination, legal threat, medical/financial claim, employee accusation, or anything you cannot verify from records.
Reply templates
Positive review
Thank you for taking the time to share this. We appreciate the chance to help with [verified service]. If you ever need anything else, the best way to reach us is [official contact path].
Neutral or mixed review
Thanks for the feedback. We are glad we could help with [verified part] and we are reviewing the note about [general issue]. If you are open to it, please contact [official path] so our team can look at the details from your service record.
Negative review with uncertain facts
Thank you for flagging this. We do not want to guess at private account details in a public reply, but we do want to review what happened. Please contact [official path] with [non-sensitive identifier] and a manager will check the record.
Spam or wrong business suspected
We are not able to match this review to a customer record from the information shown here. If this was meant for our team, please contact [official path] so we can review it. If not, this may have been posted to the wrong business profile.
AI drafting prompt
You are helping draft a public review reply for a small business. Use only the facts below. Do not invent customer identity, service details, diagnoses, pricing, refunds, warranty coverage, timelines, or private account information.
Business tone: calm, helpful, concise, no arguments.
Review sentiment: [positive/neutral/negative]
Verified facts we can mention: [facts]
Facts we must not mention or invent: [facts]
Official private follow-up path: [email/phone/form]
Escalation rule: If the review involves refund, warranty, safety, legal, discrimination, or a fact not in the verified facts list, output NEEDS OWNER REVIEW instead of a public reply.
Draft one public reply under 90 words. Include the private follow-up path only when useful. Do not make promises beyond the verified policy.
Five-minute review before posting
- Can we prove every service detail in the reply from a customer record?
- Does the reply avoid private personal information?
- Does it avoid admitting fault, refund approval, warranty coverage, or diagnosis unless approved?
- Does it point unhappy customers to the official private follow-up path?
- Would the reply still be safe if quoted by Google AI Overviews, ChatGPT Search, Perplexity, Gemini, or Copilot?
Related Horizon Flow resources
For AI-search website proof and trust signals, see SEO After AI. For local-service lead handling and customer follow-up scripts, see Local Lead Rescue System.
Prefer a free starting point? Use the AI prompt governance checklist before saving reusable review-response prompts.