Google Review Automation for HVAC Contractors
One HVAC contractor quadrupled monthly Google reviews in 90 days with an automated post-job SMS — 35% of customers responded. Here is the exact flow, the timing, and how to stay inside Google's rules.
An automated post-job review request — sent by SMS within 24 hours of job completion, with one follow-up for non-responders — got a 35% response rate and quadrupled monthly Google review volume within 90 days for an HVAC contractor completing 40–50 jobs a month.
No incentives, no gating, no begging. The whole system is timing and consistency, and it's automatable end to end.
Why review counts stall at most shops
Your techs are supposed to ask at the door. They forget, or it feels awkward after collecting payment, or they ask and the customer says “sure” and never does it. Asking depends on a human remembering at the exact moment they most want to leave — so it doesn't happen.
Meanwhile reviews are the local pack's currency: when a homeowner searches “AC repair near me,” volume, recency, and rating decide who shows up. The competitor with 300 reviews isn't better than you; they're just systematic about asking.
The flow that got 35% response
| Step | When | What goes out | Why it works |
|---|---|---|---|
| Job closed in the system | Trigger | Nothing yet — the trigger is the tech marking the job complete | No manual step; the automation rides your existing workflow |
| Review request SMS | Within 24 hours | Short personal text naming the tech and the job, with a direct Google review link | Sent while the fixed AC is still a relief, not a memory |
| Follow-up SMS | 2–3 days later, non-responders only | One gentle nudge, same link | Roughly a third of eventual reviews come from the nudge |
| Stop | After one follow-up | Nothing further | Two touches is persistence; three is pestering |
Staying inside Google's rules
Automating the ask is fine. What violates Google's policy is review gating — asking only happy customers, or pre-screening sentiment and routing unhappy ones away from Google — and offering incentives for reviews. Send the same request to every completed job and you're clean.
That policy constraint is also why honest automation beats clever automation: a steady stream of authentic reviews, including an occasional imperfect one, reads as real to both Google and homeowners. A wall of five-star reviews that all arrived in one week does not.
What about the customer who had a bad experience?
They get the same request — and occasionally you'll get a rough review. Treat it as free quality control with a public reply opportunity: respond fast, own the miss, offer to make it right. A 4.8 average with thoughtful owner responses converts better than a suspicious 5.0.
The operational win is that you hear about problems within 24 hours of the job instead of weeks later, while it's still fixable and before it becomes a one-star story.
What owners ask about review automation
How do I get customers to leave reviews without begging?
Ask every customer, once, by text, within 24 hours, with the direct link — and have a system do it so it actually happens. The published numbers: 35% of customers respond to a well-timed SMS request with one follow-up. Begging feels necessary only when asking is rare and random; at 40–50 jobs a month, systematic asking produces 15–20 new reviews monthly.
Will automated review requests get me flagged by Google?
Automation isn't the violation — gating and incentives are. Google prohibits asking only happy customers, screening sentiment before showing the review link, and paying or discounting for reviews. Send an identical request to every completed job and automation is just consistency. That's exactly how the deployment in our case study runs.
When is the best time to ask for a review?
Within 24 hours of job completion — ideally the same evening. The customer's relief is freshest right after the fix, and the tech's name still means something. Wait a week and the job is a forgotten line item. This timing is the single biggest lever in the whole system, and it's the thing manual asking always gets wrong.
What response rate should I actually expect?
The published deployment sustained 35% — roughly one review per three completed jobs — using one request plus one follow-up. If you're getting under 10%, the usual culprits are asking by email instead of SMS, asking days too late, or a link that takes more than one tap to reach the review box.
Should I reply to every review, including the bad ones?
Yes, and prioritize the bad ones — reply within a day, own what went wrong, and take the fix offline. Prospective customers read your worst reviews first, and what they're actually evaluating is your response. A well-handled one-star review is some of the best marketing on your profile.
The shift
Reviews are a systems problem wearing a marketing costume: trigger, timing, follow-up, stop. Review automation is one of the four AI systems Astola builds for HVAC contractors, and it usually ships alongside call coverage.
Want this running on your jobs board?
Book a free call. If your job-completion data lives anywhere structured — ServiceTitan, Housecall Pro, GHL, even a spreadsheet — this system typically goes live within a week.
Book a free discovery callM. Hasan Tariq
M. Hasan Tariq is the founder of Astola Consulting, an AI consultancy that builds custom AI Operating Systems for busy entrepreneurs. Before Astola, Hasan spent years in enterprise DevOps and AI automation at Systems Limited. He works with boutique agencies, consultants, and operators who need modern solutions to their modern problems.