Getting more reviews, legally: the complete guide

Last fall a flooring contractor in Grand Rapids forwarded me a pitch from a review-generation agency. For $249 a month, their software would text every customer after a job: “How would you rate your experience, 1 to 10?” Nines and tens got a Google review link. Everyone else got routed to a private feedback form that the owner would see and Google never would. The pitch deck called this “reputation protection.”

It’s review gating. Google’s contributed-content policy prohibits it by name, and Google blocked or removed 292 million policy-violating reviews in 2025, up 21% from the year before (reported April 2026). The agency didn’t mention that. They rarely do, because most of what the review-generation industry sells violates Google policy, the FTC’s review rule, or both. The playbook that survives both rulebooks is smaller than the gurus claim, and it works better, because a clean review base compounds while a gamed one sits one policy sweep away from zero.

This post is the full map: what each rulebook actually prohibits, which popular tactics fail which rulebook, the asks that pass both, and what to do about the gray zones everyone argues about at marketing conferences.

Two rulebooks, and they don’t agree

The single biggest source of confusion in this space is that people talk about “the rules” as if there were one set. There are two, written by different institutions for different reasons, and they overlap without matching.

Rulebook one: the FTC, which cares about fraud and money

Since October 21, 2024, the FTC’s Rule on the Use of Consumer Reviews and Testimonials (16 CFR Part 465) has made a specific list of practices illegal in the United States. The list is worth reading in the original, and the FTC’s own Q&A on the rule (2024) is unusually readable for a federal document. The short version of what it bans:

• Fake reviews, including AI-generated ones, from people who never used the product.
• Buying or selling reviews.
• Compensation that’s conditioned on the review’s sentiment, positive or negative. Paying for a review at all is a problem; paying for a five-star review is the version the FTC named explicitly.
• Undisclosed insider reviews: officers, employees, and their relatives reviewing the business without saying so.
• Suppressing honest negative reviews with unfounded legal threats.
• Presenting a displayed set of reviews as representative when you’ve suppressed the negative ones.

Each violation carries a civil penalty of up to $53,088 (the 2025 inflation adjustment, still the figure through 2026). Per violation. A campaign that generates forty fake reviews is not one problem, it’s forty.

Notice the theme: the FTC rule is about deception and money changing hands. It is a fraud statute wearing a marketing hat.

Rulebook two: Google, which cares about its own data

Google’s contribution policy is stricter than the FTC in a different direction. Google bans:

• Review gating: discouraging negative reviews or selectively soliciting positive ones. The funnel from the intro, in other words.
• Incentives for reviews, regardless of sentiment. Payment, discounts, free goods, and incentives to revise or remove a review are all prohibited, even if you’d happily reward a 1-star the same as a 5-star.
• As of April 2026: review quotas for staff, and asking customers to include specific content in a review, such as an employee’s name.

Here’s the distinction to tattoo somewhere visible, because the gurus blur it in both directions. The FTC rule does not flatly ban selective solicitation. Google’s policy does. The closest the federal rule gets to gating is that representativeness clause: if you suppress negatives and then present what’s displayed as the honest picture, you’ve crossed into deception. But a business that quietly asks only its happiest customers for reviews hasn’t automatically broken federal law. It has, however, violated the policy of the platform hosting nearly all of its reviews, which for most local businesses is the more dangerous of the two.

And the inverse: Google’s incentive ban is total, while the FTC’s sharpest incentive language targets compensation tied to sentiment. So “the FTC allows it” and “you can do it” are different sentences. More on that in the gray-zone section.

Who enforces what, and how it actually feels

Enforcement is where the two rulebooks stop being abstract.

The FTC is loud and slow. In December 2024 it reached a $20 million proposed judgment against Leader Automotive, a dealer group whose employees were required to post fake positive reviews. In December 2025 it took its first enforcement step under the new rule specifically: warning letters to ten unnamed companies for things like paying employees to have friends and family post 5-star reviews, and soliciting reviews from people who’d never used the product. As of mid-2026, no company has been publicly fined or settled under the rule itself. That will change, and when it does the number attached will make trade press headlines. (We’re tracking the rule’s rollout in our review of the rule’s first year.)

Google is quiet and fast. No press release, no warning letter. Your review count drops, or new reviews stop posting, or a chunk of your profile’s history evaporates during one of the sweeps that removed those 292 million reviews in 2025. There’s no docket number to look up and often no notification at all.

Think of it like a building inspector and a landlord. The inspector (the FTC) shows up rarely, and when he does the fine is enormous and public. The landlord (Google) can just change the locks, this afternoon, without telling you why. Most businesses obsess over the inspector and get evicted by the landlord.

The standard playbook, scored against both rulebooks

Run the common “review generation” tactics through both filters and watch how few survive.

Buying reviews from a vendor. Fails both, catastrophically. This is the core of what the FTC rule was written to kill, and it’s the easiest pattern for Google’s detection systems to catch: bursts of reviews from accounts with no local history. If an agency promises a specific number of reviews per month, ask exactly where those reviews come from before you sign anything. The answer is usually the whole story.

Discounts or freebies for reviews. Google bans this outright, sentiment or no sentiment. The FTC gets interested the moment the reward is tied to what the review says, and “show us your 5-star review for 10% off” is exactly that. There’s a wide belief that a small discount for “an honest review” threads the needle. It threads one needle and impales itself on the other.

The sentiment funnel. The $249-a-month product from the intro. Legal under federal law in most configurations, prohibited by the platform where the reviews live. This is the tactic with the widest gap between how it’s marketed (“industry standard”) and its actual policy status.

Employee and friends-and-family drives. Under the FTC rule, insider reviews without disclosure are a named violation, and “have your cousin post something nice” is what two of those December 2025 warning letters were about. On the Google side, the April 2026 policy update added staff review quotas to the banned list. If your team huddle includes a reviews-per-tech target, that target is now a policy violation. (The employee question has enough wrinkles that we gave it its own post.)

Pre-writing the review for the customer. Handing a customer AI-drafted text to paste in, or asking them to mention your technician Dave by name so his profile ranks for “Dave,” both now sit on the wrong side of the line. The FTC treats reviews that don’t reflect the reviewer’s actual experience as fake regardless of who typed them, and the named-employee ask is explicitly out under Google’s April 2026 additions.

That’s the guru playbook. Roughly all of it.

What survives: the boring playbook that compounds

Here’s what’s left when you filter for “passes both rulebooks,” and why it outperforms the gray stuff anyway.

Ask everyone, the same way. Every customer gets the same ask, regardless of how the job went. This is the whole compliance core in one sentence: no pre-screening, no sentiment filter, no quiet list of customers you skip. It also produces the thing prospects actually trust, which is a review page with texture. BrightLocal’s 2026 survey found 97% of consumers read reviews for local businesses; those readers have seen a thousand suspiciously flawless profiles and they discount them on sight.

Make it one tap. A direct link to your Google review form, sent by text or email. Not a link to your website, not “find us on Google.” Every additional step loses people who were fully willing ten seconds earlier.

Ask at the moment of expressed satisfaction. Not on a schedule. When the customer says the floor looks great, when they email a thank-you, when they tell your receptionist the visit was easy: that’s the window. More on this below, because it’s the highest-leverage variable in the whole system.

Follow up exactly once. One reminder, three to five days later, then stop. A second reminder buys you almost nothing and starts costing goodwill.

Respond to what you get. Solicitation and response are one loop, not two programs. Customers check whether reviews get answered before deciding whether writing one is worth their time, and prospects read your responses as a sample of how you’ll treat them. Responding to praise is the easier half to systematize (there’s a template set for positive review responses if you want a starting point that isn’t “Thanks so much!!” forty times in a row). The harder half, negative reviews, has a full playbook of its own.

Five moves. None of them clever. The reason they beat the guru playbook isn’t ethics, it’s arithmetic: reviews earned this way never get clawed back in a sweep, they arrive in a steady drip instead of suspicious bursts, and the drip is exactly what the market now rewards. In that same 2026 BrightLocal survey, 74% of consumers said they want to see reviews from the last three months and 32% want them from the last two weeks. Recency now weighs about as heavily as rating. A gamed 4.9 with nothing new since March loses to an honest 4.7 that got three reviews this week.

What this looked like for one business

Back to the Grand Rapids flooring contractor, Callister & Sons. He didn’t sign with the funnel agency. Instead we built the boring version: at the final walkthrough, whichever crew lead ran the job makes the ask in person, and only after the customer has reacted to the finished floor. The script is two sentences: “If you’re happy with how it turned out, a Google review honestly helps us more than any ad we run. I’ll text you the link so you don’t have to go looking.” The text goes out within the hour with the direct review URL. One reminder on day four. That’s the entire system.

In the six months prior, Callister had collected 11 reviews. In the six months after, 47. Two of them were 3-star reviews that a funnel would have intercepted, and his average slid from 4.9 to 4.8. Nobody cared. Calls from his Google Business Profile were up 31% over the same stretch, which he attributes mostly to being the only flooring company in his area whose most recent review was days old instead of months. The 3-stars got thoughtful public responses, and he’s convinced the pair (complaint plus answer) sells better than a spotless page ever did.

The timing of the ask beats the volume of asks

Now the part that cuts against how nearly every review tool is built.

The industry optimizes send volume: more requests, more automation, more drip sequences, on the theory that review count is a function of ask count. It mostly isn’t. Ask conversion is a function of when the ask lands relative to the customer’s peak of satisfaction, and that peak is an event, not a date. It’s the moment they compliment the work. An ask that arrives inside that window converts at multiples of the identical ask sent by a scheduler 72 hours later, when the floor is now just the floor.

This is why the crew-lead script above outperforms automation that sends five times as many requests. It’s also why the best “review software” setup for many small businesses is a human trigger (the person who heard the compliment hits send) attached to an automated delivery.

And it reframes what those funnel products are actually selling. Strip away the dashboard and the “1 to 10” pre-screen is the product: a sentiment filter that exists to keep unhappy customers away from Google. That’s gating with extra steps and a monthly invoice. The delivery mechanics bundled around it (the link, the reminder) are the legal part, and you can run those yourself for approximately free. If a tool asks the customer how they feel before showing them the review link, it’s a gating machine, whatever the pricing page calls it.

The gray zones: charity, raffles, and “technically not paying them”

Every discussion of review incentives eventually produces the same two workarounds. We’ll donate $5 to charity for every review. Or: every reviewer this month is entered to win a $200 gift card. The reasoning is that no individual reviewer gets compensated, or the compensation isn’t conditioned on sentiment, so the FTC’s sharpest language arguably doesn’t reach it. And you can, in fact, build a respectable-sounding argument that a donation in someone else’s name isn’t compensation at all.

Here’s the honest answer: the argument doesn’t matter, because Google’s incentive ban doesn’t have a sentiment carve-out or a charity carve-out. Payment, discounts, free goods, incentives of any kind, for posting or for revising or for removing: banned regardless of what the review says. A raffle entry is an incentive. A donation triggered by your review is an incentive. The gray zone is only gray in the rulebook that doesn’t host your reviews. The rulebook that does host them is black and white, and its enforcement mechanism is deleting the thing you were trying to build.

So the practical rule for incentives is one line: nothing of value, from anyone, for any review, in any direction. If you’re weighing a specific scheme (points programs, loyalty perks, “review us and we’ll waive the fee”), the breakdown of discount-for-review schemes walks through the variants one by one. Spoiler: the variants mostly end the same way.

The part nobody tells you

The real penalty for running the gray playbook isn’t the $53,088 figure, which so far no one has actually been made to pay under the rule. It’s that platform enforcement is retroactive and indiscriminate. When Google acts on a profile, it doesn’t carefully separate the reviews your funnel produced from the ones you earned; suspicious patterns can take legitimate reviews down with them, and the appeal path is a single shot with no second tier. Three years of accumulated social proof is the most valuable marketing asset most local businesses own, and it’s the only one that can’t be repurchased at any price. That asymmetry, not virtue, is the business case for the clean playbook.

If you’re currently running a funnel, the fix is smaller than you’d think: keep the send, delete the pre-screen. The link, the timing, the single reminder all stay. The only thing you remove is the question that decides who deserves to see the link. And if you inherited a profile with a gamed history, stop adding to it now; a clean recent stream is also how you dilute an ugly past, since consumers weight the last three months so heavily anyway. About 68% of consumers won’t use a business rated below 4 stars (BrightLocal, 2026), and the path back above that line is volume of honest recent reviews, not surgery on old ones.

Getting the ask right is one leg of a bigger system, alongside monitoring and response; the reputation management guide for small businesses covers how the legs fit together. And when you’re ready to implement, the companion piece on asking for reviews without violating policy has the word-for-word scripts, send timing, and platform-by-platform details.

The gurus are selling speed, and speed is what gets profiles wiped. The two rulebooks, read together, leave you a playbook with five boring moves and one enormous advantage: everything it builds, you get to keep.