Every review answered, in your voice, the same day
AI watches every review platform, drafts a response in your tone for each one, and queues anything negative for a human decision before it goes out.
The problem
How it works by hand
For local and consumer-facing businesses, Google reviews are the shop window, and a wall of unanswered reviews reads as not caring. But responding properly means checking several platforms, remembering who the customer was, and writing something that is not the same thank-you pasted forty times. So it slips, and the one-star review sits there unanswered for a month, doing damage daily.
A worked example
What a working version looks like
The system monitors every platform you are reviewed on, Google, Facebook, Trustpilot, the industry-specific ones, and picks up new reviews within minutes of them appearing. For each one it drafts a response in your tone: specific to what the reviewer said, varied so your profile does not read like a mail merge, and matched against your booking or order records where possible so the reply can reference the actual visit. Positive reviews can go out automatically or via approval, your choice. Negative reviews never auto-send: they are queued with a drafted response and the customer’s history, for a human to approve, soften or rewrite. Everything is logged, so you can see response coverage across every platform in one place.
The exact tools change per business. The shape does not.
What it needs
Honest inputs, nothing exotic
- 01Access to your review profiles (Google Business Profile at minimum)
- 02A feel for your tone: a few past responses you were happy with is enough
- 03A named person to approve the negative ones
- 04Optionally, your booking or order system for matching reviewers to visits
The payoff
What you get back
Reputation upkeep becomes a five-minute approval job instead of a guilty backlog. Every review gets a same-day response that sounds like you, and the negative ones get handled deliberately rather than discovered late.
Do it yourself
How you would build this yourself
No course, no upsell. This is the order we would build it in, with the tools named, and a prompt to start from.
- 1
Switch the notifications on: Google Business Profile, Facebook and Trustpilot will all email you when a review lands. Free and instant.
- 2
Pick three or four past responses you were happy with. That is the tone sample, and it is all the AI needs.
- 3
For each new review, give Claude the review plus the tone samples and ask for a response that addresses what the reviewer actually said. A profile of identical thank-yous reads worse than silence.
- 4
Automate the boring half: Zapier or Make can catch new Google reviews and queue drafted responses. Positive ones post on approval, negative ones always wait for you.
- 5
Where you can, match reviewers to your booking or order records so the reply references the actual visit rather than nothing in particular.
Build me a review response system. Watch my Google Business Profile for new reviews via the API. For each one, draft a response in my tone (four examples attached) that responds to what the reviewer specifically said, no copy-paste gratitude, and vary the phrasing across drafts so the profile does not read like a mail merge. Positive reviews: queue for one-click approval. Three stars or below: never auto-respond, send it to me with the draft and the customer’s booking history from [system] if there is a match.
Copy it into Claude Code, fill the brackets, and it will plan the build with you before writing a line of code.
We would rather show you how than bill you. The whole ladder of free help, answers, guides and the weekly build-along, is on the do-it-yourself page.
Go deeper