Reply to a negative review
Professional structure: role, context, objective, details, steps, format, criteria and self-check. Designed for better results, not guaranteed ones: always review the answer.
Never use real personal data. If you need to, run it through the data cleaner.
## Role Act as an online reputation manager. ## Context A customer has left a negative review and I want to respond well. ## Objective Write a public reply that shows professionalism to future customers. ## Details (fill in what is in square brackets) - [review, without the customer's name] - [what happened according to our records] - [solution we can offer] ## Step-by-step instructions 1. Identify the main complaint and whether it is partly fair. 2. Thank and acknowledge without admitting blame that is not yours. 3. Explain briefly and offer to continue through a private channel. 4. Check there is no personal data or defensive tone. ## Output format A 60 to 100-word reply and a shorter version. ## Quality criteria - It does not argue with or blame the customer. - It reveals no customer data. - It offers a specific way to get in touch. ## Self-check Before answering, check every quality criterion. If information is missing, ask me instead of making it up, and at the end state what you have assumed.
Try it here
Open model · not frontier
This tester runs on Gemma 4 26B (Google), an open-weights model served by Cloudflare. It is not a frontier model like the paid versions of ChatGPT, Gemini or Claude: it is faster and more modest, and it will sometimes get things wrong. It shows how a good prompt behaves, not what the most advanced AI can do. Paste the prompt into your usual assistant for its best version.
Fill in the details and try the prompt with an open model without leaving the page.
Objective
Write a public reply that shows professionalism to future customers.
When to use it
This prompt is essential when managing your online reputation, particularly after receiving a critical review that threatens to damage your brand's image. It works best for service-based businesses like restaurants, consultancies, or retail shops where public perception is key. However, avoid using this for trolls, spam, or abusive comments that violate platform policies; those should be reported directly to the site administrator rather than engaged with publicly. This tool helps you maintain composure and professional distance when emotions run high.
Worked example: input (fictional data)
Worked example: full response
Variants
Short and punchy
Rewrite the response to be extremely concise, focusing only on the apology and the contact request. Remove any lengthy explanations about the internal cause of the issue to keep the response under 50 words, perfect for mobile readers who want a quick acknowledgment without unnecessary fluff or excuses.
Luxury service tone
Adapt the tone to be highly formal and sophisticated, suitable for a high-end boutique or luxury hotel. Use polite, elevated language that emphasises exclusivity and personal care, ensuring the response sounds like a direct, dignified letter from the management team rather than a standard customer service template.
Internal team follow-up
Instead of a public response, draft this as an internal memo to your staff. Explain how to handle this specific type of complaint when it happens face-to-face in the venue, focusing on de-escalation techniques and when to involve a manager to ensure better results on the floor.
Common mistakes
- Using the same reply for every review.
- Arguing about the facts in public.
- Copy-pasting generic corporate jargon that sounds robotic and insincere. How to avoid it: Personalise the opening and closing sentences to show you have actually read their specific complaint.
- Apologising for things that were not your business's fault. How to avoid it: Acknowledge the customer's feelings without taking blame for factors outside your control, like traffic or weather.
- Writing a response that is far too long and rambling. How to avoid it: Keep the message tight; brevity shows you respect the customer's time and are focused on solutions.
- Using passive-aggressive language disguised as professional politeness. How to avoid it: Review your draft for hidden jabs or sarcastic phrasing that might escalate the situation further.
Related guide
- AI for hospitality: translated menus, reviews and rotas (without gambling on allergens)
- AI for small shops: product listings, photos and customer service
General and assistant-specific tips
Getting the most out of it
- Paste the prompt as is and only replace what is in [brackets].
- If the answer does not fit, do not start over: ask for “shorter”, “more formal” or “as a table”.
- Always review the result before using it: AI can get facts, dates or names wrong.
Before you paste it: quick checklist
- Don’t paste real customer data or full invoices.
- Describe your business in one sentence: trade, type of customer and area.
- Ask for gaps in [brackets] so you can fill in real data yourself.
- Check any tax or legal obligation on the relevant official website.
How to adapt it to your assistant
The template works the same in ChatGPT, Gemini, Copilot or Claude because it only uses plain language. If your assistant supports fixed instructions (custom instructions, Gems or projects), you can store your preferred tone and format there and paste only the variable part. On a free plan, split long requests into two or three shorter messages.