Per-business brand voice
Set rules once per business: tone, signature line, do's, and don'ts.
Meet the AI assistant built into SyncMyReviews. Generate an on-brand reply with one click while you read a review, or flip auto-reply on per business and the assistant posts within 5 minutes — using the tone, signature, and do/don't list you defined.
"Honestly the best service I've had in years. The team explained everything clearly, finished early, and the price was exactly what we agreed. Already booked them again for the spring."
Thank you so much, Jamie — this made our whole team's week. We're really glad the work came in on time and on budget, and we're already looking forward to seeing you in the spring. Anything we can do in the meantime, just give us a call.
Google's own ranking signals reward businesses that respond to reviews — every reply tells the algorithm (and the next prospect reading) that this is a business that listens. But replying to every 5-star "Loved it!" and every 2-star complaint, in your tone, in the reviewer's language, within minutes of the review going live — that's a part-time job nobody on your team has time for.
So most owners do one of three things:
The AI assistant fixes all three.
SyncMyReviews already syncs every new Google and Facebook review in real time through official webhooks. The moment a review is created, the assistant is notified.
The assistant reads the text, rating, your business profile, and brand-voice notes. It picks the right tone and writes a 2–4 sentence reply in the customer's language.
The reply is posted directly to Google or Facebook after a built-in 5-minute grace period — so you can still jump in if you want. If you do, the assistant quietly stands down.
Set rules once per business: tone, signature line, do's, and don'ts.
Picks a reply template based on the rating: positive (4–5), neutral (3), or negative (1–2). You don't have to think about it.
Flip a switch to enable auto-reply. Leave it off, and the assistant becomes your one-click drafting helper on the Reviews page.
Auto-reply waits 5 minutes. If you reply by hand in that window, the assistant aborts silently — no double replies, no awkward overlap.
Hard-coded to never mention being AI, never use emojis or markdown, and never invent facts. It writes like a calm business owner.
For bare 5-star ratings without text, it picks from a varied hand-written pool of “thank yous” so 50 reviews never look copy-pasted.
If the review is in Spanish, the reply is in Spanish. French → French. Language is inferred automatically; no toggles to manage.
Every AI reply is logged: business, review, tone, exact model, and prompt hash. You always have a record of what was sent and why.
Reviewer text is sanitized before it reaches the AI — zero-width characters and overrides are stripped. The review is wrapped in a strict trust boundary the model cannot break.
If the model returns a leaked instruction or known jailbreak marker, the reply is discarded and replaced with a safe canned line. You and the reviewer never see the broken output.
AI-generated replies go through the same posting path as a reply you'd type yourself — same validation, retries, and tenant guard. No privileged backdoor.
Manual generation is capped at 60 requests per hour per user. Auto-reply is one job per review, period — multiple webhook deliveries collapse into a single attempt.
Auto-reply is off by default everywhere. Members can only toggle auto-reply for businesses they have explicit access to.
"Honestly the best service I've had in years. The team explained everything clearly, finished early, and the price was exactly what we agreed. Already booked them again for the spring."
Thank you so much, Jamie — this made our whole team's week. We're really glad the work came in on time and on budget, and we're already looking forward to seeing you in the spring. Anything we can do in the meantime, just give us a call.
"Warm and professional. Mention we offer evening appointments."
/settings Open the AI Assistant card. One row per business you can access. Toggle Auto-reply on for the businesses you want covered, then teach it your tone with Brand voice notes. Saves as you type — there's no submit button.
"The team made the whole appointment so easy — booked online, in and out in 30 minutes. Couldn't recommend more."
/reviews On every Google and Facebook review you'll find Generate Response next to the regular Reply. Click, the assistant drafts a reply, you edit if you want, then post. That's the whole product surface — no extra dashboards, no workflow builder, no template library to maintain.
You run 7 dental clinics, each with its own Google listing and brand voice. Set the persona notes once per clinic, turn auto-reply on for each, and stop coordinating reply shifts across managers.
You're a one-person agency or studio. Auto-reply takes the praise off your plate so you can read 1- and 2-star reviews carefully and respond personally — the assistant still drafts those for you, you just choose whether to send.
Members on your team only see the businesses you've granted them access to. The assistant respects that scope — they can toggle and tune auto-reply for their assigned businesses, and the rest of your portfolio stays invisible to them.
There's no add-on, no per-reply charge, no usage-based pricing surprise. The AI-Powered Review Response feature is part of SyncMyReviews — same Stripe subscription, same 14-day trial, same one-click cancel from the Customer Portal.
Google Business Profile and Facebook Pages. Manual reviews, LinkedIn, and Yelp are not covered (yet) — the AI assistant only replies on the platforms where SyncMyReviews holds an official posting connection.
No — and that's a deliberate safety net. Auto-reply waits 5 minutes after a new review before posting. If you reply by hand in that window, the assistant silently steps aside.
In manual mode, yes — that's the whole point of the Generate Response button. The assistant drafts, you edit, you post. In auto-reply mode, no — the whole value is that you don't have to be in the loop. If you want to edit, leave auto-reply off for that business.
Whatever language the reviewer used. The language is inferred from the review text itself. There's no setting to manage.
The system is built to assume every review is potentially hostile. Reviewer text is sanitized (control characters, zero-width characters, length cap), wrapped in a strict trust boundary in the prompt, and the output is scanned for known jailbreak markers. If anything looks off, the assistant replaces the reply with a safe canned line.
Not from anything the assistant writes. It never says “as an AI”, never uses emojis, never uses markdown, and never adds anything to the reply that would identify it as automated.
Google Gemini, via the official Gemini API. The exact model identifier is recorded with every reply so the audit trail is precise.
Yes. Flip the switch on the /settings page back to off for any business, any time. Active auto-replies that haven't fired yet during the 5-minute window will detect the change and abort.
It's included in your SyncMyReviews subscription. No per-reply cost, no add-on tier.
Yes — every call lands in an internal audit table with the business id, review id, tone, model used, and prompt fingerprint. Nothing is invisible.
Every unanswered 5-star review is a missed ranking signal. Every unanswered 1-star review is a chance to recover a customer you've already lost. The AI assistant handles both — in your voice, on your schedule.
14 days free · No credit card required · Cancel from your account in two clicks.