Replying to reviews so the next customer trusts you
A generated draft gets you most of the way. The rest is knowing what separates a reply that wins people back from one that makes things worse, and where the real fix for bad reviews lives.
Answering a one-star review
Same angry review, two very different replies. The first is defensive and forgettable. The second acknowledges the experience, apologizes without excuses, and moves the fix offline. Every future customer reading it sees a business that takes problems seriously.
The review
“Waited 40 minutes for a cold meal and nobody apologized. Won't be back.”
A defensive reply
“We were extremely busy that night and do our best. Sorry you feel that way.”
A reply worth posting
Hi [name], I'm sorry — a 40-minute wait for a cold meal is on us, not you, and we should have said so at the table. I'd genuinely like to make it right. Could you reach me directly at [email]? I want to understand what went wrong that night and fix it for next time.
Six habits behind a reply that lands
A generated draft gets you about 80% of the way. These are the habits that close the gap, on a five star and a one star alike.
Most people who read your reply will never have left a review. They're deciding whether to trust you, so a calm, specific reply to a bad review can win more customers than the review lost.
Echo the actual detail they mentioned. "Thanks for the kind words" pasted under every five star reads like a bot. "Glad the late checkout helped" reads like a person.
On a negative review, own it in the first line. "We were slammed" is an excuse; "that wait was on us" is an apology. Never get defensive in public.
Acknowledge publicly, then invite them to a real person by email or phone. The public reply shows you care; the private channel fixes it.
Yelp bans offering freebies in a reply; app stores strip links; Google rewards a natural mention of your business name. The wrong move can get a reply removed.
It's public and permanent. AI sounds confident even when it's wrong about your details. You're the editor on every reply, not the rubber stamp.
Here's the part a response generator can't do: most one-star reviews start as a support issue that never got resolved in time. Answer customers fast enough and the frustration never hardens into a public review. That means a findable help center and an AI agent that resolves the common questions before they become complaints.
A generator vs stopping the review
A review response generator drafts a great reply after the review lands. A real AI agent works upstream: it resolves the issue while it's still a support question, before the customer ever reaches for the one-star button. Here's where the two diverge.
| Capability | Generator | Selvo agent |
|---|---|---|
| Writing the reply | ||
| Drafts a public reply to a review | ||
| Knows your product and policies | Only what you paste in | Trained on your help center |
| Applies your brand voice | Re-describe it every time | Set once |
| Stopping the bad review | ||
| Resolves the issue before it becomes a review | ||
| Answers customers 24/7 on its own | ||
| Lives inside your support inbox and widget | ||
| Hands off to a human when unsure | ||
| Trust & cost | ||
| Cites a source the customer can verify | ||
| Audit log of every answer | ||
| Price | Free | $0.10 per resolved chat |
