What Google's Review Policy Actually Prohibits (And What Most Agencies Are Still Doing)

A client calls on a Monday. Forty reviews are gone and there is a notice on their Business Profile telling customers that Google removed fake reviews from this listing.

Nobody bought reviews. What happened was a staff incentive program, a survey that filtered unhappy customers before they reached Google, and a request script asking people to mention their technician by name. Three tactics that were standard practice two years ago and are now explicit violations. Anyone handling review management for client profiles is working under a different set of rules than the ones those programs were built for.

The google review policy was updated in April 2026, and most review programs running today were built before that. This is what the rules actually say, which practices they prohibit, and what happens to a business that gets caught.

Google's Review Rules Live in Three Documents, Not One

Most articles on this subject treat the rules as a single page. They are not, and knowing the structure matters when you are trying to work out whether a specific tactic is permitted.

The parent document is the Maps user-generated content policy, which covers everything customers contribute: reviews, ratings, photos, edits. Underneath it sits Prohibited and restricted content, where the specifics live. Inside that page, one section governs almost every review question an agency will have: Fake and Misleading Content and Behavior.

That section splits into two subsections worth knowing by name.

Fake engagement covers content that does not reflect a real experience. Paid reviews, reviews given in exchange for anything of value, posts submitted from multiple accounts at one person's request, and anything pushed through an emulator or tampered device.

Rating manipulation covers the ways a business influences what gets written and by whom. This is the subsection Google changed in April 2026, and it is the one that catches agencies running structured review programs.

One point of precision worth carrying into client conversations. Most of the industry says "rating manipulation policy" as though it were a standalone document. It is a subsection. If you are citing it to a client, cite the parent policy.

Rating Manipulation Now Covers Staff-Directed Requests

On April 17, 2026, Google added two prohibitions to the rating manipulation section. Amy Toman, a Google Diamond Product Expert, spotted the change and shared it publicly, and PPC Land documented the exact wording.

The policy now bars merchants from "requesting that staff solicit a certain number of reviews." It also bars merchants from requesting that staff solicit reviews containing specific content, including content that identifies a staff member.

Read plainly, that ends four things. Review contests where employees compete on volume. Staff leaderboards tied to mention counts. Bonus structures paying out per review. And any script telling a technician to ask a customer to name them in the text.

The last one is the most widely used and the least obviously wrong. Asking a customer to mention the person who helped them feels like good service rather than manipulation. Under the current rules it is a merchant requesting reviews with specific content, and the reviews it produces are candidates for removal.

The change applies globally to every merchant with a Maps listing. There is no regional grace period and no grandfathering of existing programs.

Review Gating Is the Violation Most Agencies Commit Without Knowing

Ask an agency owner whether they gate reviews and most will say no. Ask them how their request sequence works and roughly half will describe gating.

Review gating means soliciting reviews selectively based on the sentiment you expect to receive. It does not require deliberate suppression. If the process asks a customer how their experience went, then routes happy ones to Google and unhappy ones to a private feedback form, that is gating regardless of how it is labeled internally.

The survey-first funnel is the most common version, and it looks defensible because collecting private feedback from dissatisfied customers is genuinely useful. The problem is what happens to the Google request. It fires for one group only, which means the public rating reflects a filtered sample rather than the customer base.

Compliant collection is simpler and less comfortable. Every customer gets the same request, on the same schedule, with no branching logic based on how the job went.

Here is the part most vendors leave out. Removing the filter usually lowers the average rating in the short term, because customers who were previously routed away now have a path to the profile. A business going from a gated 4.9 to an honest 4.6 has not gotten worse. It has stopped hiding, and the rating it lands on is one that survives enforcement.

Incentives Cover Anything of Value, Not Just Cash

The rule on incentivized reviews is broader than most people assume. Google's fake engagement section prohibits reviews and ratings paid for directly or in kind, which sweeps in a lot of programs where no money changed hands.

Prohibited in practice: discounts offered in exchange for a review, free products or services, entry into a prize drawing, loyalty points, and gift cards. Anything given to the customer contingent on them posting.

The reframing attempts do not work either. Offering a discount for "feedback" and then linking to Google is still an incentive. Running a monthly draw for anyone who leaves a review is still an incentive. The policy language addresses the exchange of value, not the wording chosen to describe it.

The staff-side version is now covered explicitly by the April change. A bonus tied to review counts is a merchant requesting a certain number of reviews from staff, which is prohibited on its own terms.

What remains permitted is unglamorous and effective. Ask every customer, make the process take under thirty seconds, and follow up once if they do not respond.

Enforcement Removes Reviews and Leaves a Public Warning

Google scans contributions with machine learning for signals of suspicious activity, and enforcement is automated well before it is ever manual. A business found in violation faces a combination of consequences.

New reviews and ratings can be blocked entirely for a set period. Existing reviews can be unpublished. And the profile can display a warning message informing users that fake reviews were removed from the listing.

That third consequence is the one clients react to, and it is the one worth raising in a sales conversation. Losing reviews is invisible to customers. A notice on the profile is not. It sits in front of every person who looks the business up, and it says something worse about that business than a mediocre rating ever would.

Appeals are available. The practical catch is that the warning generally stays visible while the appeal is processing, so the reputational cost is paid before any determination is made.

There is also a downstream effect that did not exist a few years ago. Google's AI reads review text when it generates local recommendations, so a profile with restricted or removed reviews has less material feeding those answers. We covered how that mechanism works in local SEO and AI.

Recovery is slow and partial. Prevention is the whole game.

How to Audit a Review Program Against the Current Rules

Run this against any client program before the next request goes out. It takes about twenty minutes per profile.

Check whether every customer gets a request or only some of them. Any filtering step, branching survey, or manual selection means the program is gating. Remove the branch.

Look at staff compensation next. Contests, bonuses, and leaderboards tied to review counts or mentions all need to go, including the informal ones nobody wrote down.

Then check where requests happen. On-site asks, shared tablets, and review kiosks are all restricted. Move everything to email or SMS sent after the customer has left the premises.

Read the request script. Naming a service, a keyword, or an employee crosses into requesting specific content. Ask what the customer wants to say and stop there.

Confirm nothing of value is attached, including non-cash items, entries, and points.

Finally, check whether the profile already carries a warning. If enforcement has already happened, the fix sequence changes and the client needs to know before anyone promises them growth.

For multi-location clients, run this per profile rather than once across the account. The 21 GBP tips agencies use covers the rest of the profile-level work that runs alongside it.

Frequently Asked Questions

Is it against Google's policy to ask customers for reviews?

No. Asking is permitted and encouraged. The restrictions apply to who you ask, what you offer them, and whether you influence what they write. A request sent to every customer with nothing attached is compliant.

Can you offer a discount for a Google review?

No. Google prohibits reviews paid for directly or in kind, and a discount is payment in kind. The same applies to free products, gift cards, loyalty points, and prize draw entries. Rewording the offer as compensation for feedback does not change how the policy reads it.

What is review gating and why is it not allowed?

Review gating is soliciting reviews only from customers you expect to be positive, usually by surveying them first and routing the satisfied ones to Google. It is prohibited because the resulting rating reflects a filtered sample rather than the actual customer base. Any branching logic in a request sequence is worth examining for this.

Can staff ask customers for reviews?

This narrowed in April 2026. Merchants can no longer direct staff to solicit a set number of reviews, or to solicit reviews containing specific content including their own name. Staff mentioning that a request is on its way by text is safer than staff asking directly.

Why did Google remove our reviews?

Common causes are incentives, gating, staff-directed requests, requests made on a shared device, and a burst of reviews arriving in a short window from a similar source. Check whether the profile is displaying a warning, which indicates enforcement rather than routine filtering.

What happens if you violate Google's review policy?

New reviews can be blocked for a set period, existing reviews can be unpublished, and the profile can display a notice telling customers that fake reviews were removed. Appeals are possible, though the warning generally remains visible during the process. Profile-level cleanup runs alongside GBP profile optimization.

What a Compliant Review Program Looks Like

Strip out the prohibited tactics and what remains is straightforward. Requests go out on a fixed schedule after service completion, to every customer, by email or SMS. Each one carries the customer's name and the job that was done so it reads as a follow-up rather than a broadcast. None of them suggests what to write. Responses go out same day, including to the one-star reviews, because reply speed affects whether the next customer bothers writing at all.

That is less exciting than a staff contest and it will not produce a spike in month one. It also does not disappear in an enforcement sweep, which is the only version worth building a client relationship on.

This is the part most agencies struggle to staff, because it sits between customer service, compliance, and copywriting. Dialed-In Web runs white label review management for agency partners as part of a wider white label GBP SEO program, built to survive the rules as they are now rather than as they were. If you want to audit one client profile before committing to anything, start a conversation about your market or call +1 (214) 307-2995.

BP #2