HVAC Review Generation: A System That Stays FTC-Legal

All rows are BrightLocal LCRS 2026 except the last (Scorpion 2026); "not published" means no prior-year figure was printed.

The share demanding 4.5 stars or better nearly doubled in twelve months, 17% to 31%; the two-week recency share moved 20% to 32%. Recency kills a static count: 180 reviews that stopped growing in 2024 read, to the 74% who only care about the last three months, as nothing relevant.

Neither survey publishes a reviews-per-month target. Combining the two thresholds — 20 reviews as the floor, a 90-day relevance window — puts the floor near seven reviews a month. That is our arithmetic on their numbers, not a BrightLocal finding.

Ranking is a separate question from trust. Google's local ranking guidance names review count and rating as prominence inputs but no velocity or recency factor, and states there is "no way to request or pay for a better local ranking on Google." Velocity earns its place here because buyers demand recency, not because Google confirmed it.

Three HVAC job types want different triggers; a fourth wants no ask at all. Treat the table and the five steps under it as our operating defaults, not survey findings or platform rules.

Google supplies the mechanism: a review link or QR code generated from the Business Profile, which Google suggests putting on receipts, in thank-you emails, at the end of a chat interaction, or printed and displayed. QR codes can currently only be generated on a computer browser.

The five-step workflow:

1. Fire the request from job close, not from a person's memory. One request per completed job, sent by the office.
2. Send the same neutral wording to every customer. No branching on predicted sentiment, no happy-path template.
3. Keep it short and say so. "Posted in the last month" ranks third among the factors buyers weigh (44%); "long and detailed" ranks last (26%).
4. Follow up once, a few days later, then stop. One reminder, one channel.
5. Ask the technician to mention that a text is coming — nothing more. The handoff is the tech's job, not the pitch.

Step 5 is where shops get into trouble. A per-review spiff for techs named in Google reviews violates policy on its face. Google's Maps policy says "when soliciting reviews, merchants should not require or pressure users to leave ratings or write reviews while on the premises, nor should they request that specific content be included," and names two examples: "Merchants requesting that staff solicit a certain number of reviews" and "Merchants requesting that staff solicit reviews that include specific content, including content that identifies a staff member." The bonus board is the first; "mention me by name" the second. The FTC adds: "Don't ask your staff to write reviews of your business, at least not without ensuring that they disclose in their review that you employ them and asked them to write it." Somebody has to own this workflow, which is why it sits inside the review engine we run alongside map-pack work.

Google's Maps policy lists, under what merchants may not do: "Discourage or prohibit negative reviews, or selectively solicit positive reviews from customers". A pre-screen that filters unhappy customers away from your review link is selective solicitation, in Google's own words. Google permits soliciting genuine experience "without offering incentives to do so or attempting to influence the rating or the contents of the review."

The legal side is narrower than most HVAC marketing content claims. Asked whether a business may solicit only customers it thinks are happy, FTC staff answer: "The rule does not contain a specific prohibition against such conduct. But this practice could violate the FTC Act." Pre-screening is therefore not a per-se rule violation, but conduct the regulator has said may be unlawful. That rule was approved 5-0 on August 14, 2024, and the same staff Q&A dates its effect to October 21, 2024. It prohibits outright: review suppression that implies the displayed set is all or most submitted, incentives conditioned on positive sentiment, undisclosed insider reviews, and unfounded legal threats or intimidation to remove a review.

The rule "authorizes courts to impose civil penalties for knowing violations," and the ceiling is $53,088 per violation under 16 CFR 1.98(d), which applies "only to penalties assessed after January 17, 2025." The widely repeated $51,744 is superseded — the FTC announced the increase in February 2025 — but don't assume a fresh figure every January. The governing statute "directs agencies to implement annual inflation adjustments based on a prescribed formula," while 1.98's amendment history stops at that January 2025 change, with Title 16 shown current as of August 13, 2026.

The closest enforcement case is a gating case. Fashion Nova paid $4.2 million to settle FTC allegations that it suppressed reviews rated lower than four stars out of five from late-2015 until November 2019 — "the hundreds of thousands of lower-starred, more negative reviews." "Deceptive review practices cheat consumers, undercut honest businesses, and pollute online commerce," said Samuel Levine, Director of the FTC's Bureau of Consumer Protection. The FTC also sent letters to 10 companies offering review management services, "placing them on notice that avoiding the collection or publication of negative reviews violates the FTC Act." That reaches any HVAC shop whose reputation software ships a sentiment filter: "you can be held responsible for what they do on your behalf."

Section 465.7 lets you reply publicly, but FTC staff spell out narrow limits: no accusation about the reviewer you know to be false or make with reckless disregard for its truth, no physical threats, no intimidation — defined to include "abusive communications, stalking, character assassination, and sexual harassment." Legal threats are permitted only "if you have a legitimate basis for doing so." Two things the rule does not prohibit: contacting a negative reviewer to resolve the issue, and "simply asking satisfied customers to update their reviews."

The four-step protocol, whose 24-hour clock is our standard rather than a platform rule:

1. Reply inside 24 hours, publicly, in one short paragraph. Address the reviewer by name and name the issue. Google's reply guidance asks for replies that "share a helpful update or answer a question" instead of the same thank-you to everyone, and warns against offering deals in a reply.
2. Move the details to a phone call. State what you'll do, give a named person to call, and don't litigate the invoice in public.
3. Fix it, then say nothing more. Google notifies the reviewer and lets them change the review — and the date changes to the latest update, which puts a repaired review back inside the 90-day window buyers read.
4. Flag only genuine policy violations. Non-customers, competitor posts and personal information are policy matters; honest negative reviews are not.

Google screens every response before publishing, and replies "usually take up to 10 minutes to review, but sometimes a review can take up to 30 days" to appear. Verification is required before you can reply at all. Build your standard on when the reply is written, not when it appears.

A wall of five-star reviews reads as bought. Google's position: "a mix of positive and negative feedback often feels more trustworthy." BrightLocal's top-ranked factor is not the star count but whether "the review is backed up by other reviews with similar sentiment" (56%). Consistency is the asset, a handled complaint included.

BrightLocal found use of ChatGPT and other generative AI tools for local recommendations rose from 6% to 45% in a year, the third most popular source of business recommendations, while Google slipped from 83% to 71%. Reviews are read through that layer too: 82% of consumers read AI-generated review summaries and 23% would decide on the summary alone. Scorpion's home-services numbers run more conservative — 22% of homeowners use AI tools such as ChatGPT to research or find recommendations, while 83% still start online.

Three constraints keep this honest, and they are why nobody can sell an AI-citation guarantee:

• Citations churn weekly. Across 82,619 prompts over 17 weeks, SISTRIX measured weekly domain-level source churn of 5% in AI Overviews, 56% in Google AI Mode and up to 74% in ChatGPT Search [VENDOR — SISTRIX]. Read that 74% as the top of the range, not a US reading: its ChatGPT breakdown is Germany 74%, UK 60%, France 42%, with no US ChatGPT figure published — the US number, 54%, is for AI Mode. A recommendation won in week one may be gone in week two.
• A citation is not a click. Pew found users clicked a link inside a Google AI summary in 1% of visits to pages with a summary. Measure presence as presence, not as traffic.
• The strongest correlate is off-site mentions, and it is only a correlation. Ahrefs' study of 75,000 brands put branded web mentions at 0.664 Spearman correlation with AI Overview brand visibility versus 0.218 for backlinks [VENDOR — Ahrefs], while stating in-article that "correlation ≠ causation." The sample was filtered to DR>40 domains, so it describes established brands, not a new HVAC site.

Reviews remain the most defensible mention supply an HVAC company owns: real customers describing real work, where AI systems already read. Google is explicit that manufactured alternatives don't substitute — "seeking inauthentic 'mentions' across the web isn't as helpful as it might seem". BrightLocal CEO Myles Anderson puts it from the review side: "We've moved past the era where reviews were just a nice-to-have 'marketing tactic'. They've become an essential piece of evidence that your business is active, reliable."

The honest deliverable is measurement, not a promise. Tracking whether the assistants name your company, and which sources they lean on, is the job of our AI-answer tracking work for contractors. Anyone selling a guaranteed ChatGPT placement is selling against a source set SISTRIX watched replace most of itself every week.

Where we come into it: we draft the review responses ourselves rather than coaching your office staff to write them, quoted per shop. The ask itself still has to be made in person: whatever workflow sits behind it, it is your technicians, at your customers' doors, who make the request land.