Asking customers for reviews: what Google permits and what it prohibits

Google permits you to ask a customer for a review, and it documents how: a link and a QR code you generate inside your own Business Profile, plus four places it suggests you put them. What it prohibits is attaching anything to the request, dictating what the review says, and asking people whose experience was not genuine.

The permission is one sentence

Google’s Maps User Generated Content Policy spends most of its Rating Manipulation section on what a merchant may not do. One bullet states what a merchant may:

Solicit or encourage the posting of content that does represent a genuine experience, without offering incentives to do so or attempting to influence the rating or the contents of the review.

Three conditions, all inside that sentence: nothing offered, no attempt to influence the rating, none to influence the contents. Nothing in it limits how often you ask, which channel you ask on, or how you phrase it. So a plain request such as “If you have a minute, a Google review helps other people find us” is inside the permission, at the counter, on a receipt or in an email.

The link and the QR code

The mechanism has its own Google page, which opens: “To make it easier for customers to leave reviews on your Business Profile, you can create and share a link or QR code to request reviews.”

Open your Business Profile, select the Get more reviews control under Read Reviews, then copy the link or save the QR code image. The artefact itself, including what Google never publishes about its format and what happens to it when the profile changes, is in the link and where it comes from.

Google also names the channels it expects you to use: “You can also send review requests through email, WhatsApp message, or Facebook post.”

The tips page adds one condition on the customer side: “Customers must be signed into a Google Account to leave a review. They can set one up with a non-Gmail email address.” A QR code handed to somebody with no Google account produces nothing.

Where Google says to put it

The same page lists four placements:

• “Include it on your receipts”
• “Include it in thank you emails”
• “Add it at the end of a chat interaction”
• “Print and display the QR code in your store”

The chat bullet endorses the ask inside a conversation the customer already started. The in-store bullet endorses a printed sign, which is not a staff member standing next to a tablet, for the reason below.

Google’s name for it, and the rejection reason that carries it

Google’s vocabulary for an ask with something attached is fake engagement, which is also one of the reasons it can reject a reply you write. FAKE_ENGAGEMENT is a value in the PolicyViolation enum returned by the v4 Business Profile API, quoted in full:

Content that violates our policy on fake engagement isn’t allowed. Fake engagement includes incentivized content, content intended to manipulate a place’s rating, and other content that doesn’t represent a genuine experience.

It arrives in a field documented as “Optional. Output only. The policy violation that resulted in rejection. Only populated if reviewReplyState is REJECTED.” One sentence, three tests: the reward, the motive, and a catch-all for content that “doesn’t represent a genuine experience”. Google’s glossary defines the standard that catch-all is measured against:

A Genuine or Real Experience is one that reflects a user’s genuine personal experience at or with the place for which they are leaving a review.

A second glossary entry sets the width of the incentive rule:

Incentivized reviews are reviews and ratings influenced by a payment, discount, free goods or services, or any other benefit offered to the reviewer; or biased by a personal or professional conflict of interest.

“any other benefit offered to the reviewer” is the operative clause: a free coffee, a loyalty point and a prize draw entry are each inside it.

The four rules, in Google’s wording

Under “We do not allow merchants to”, the Rating Manipulation section prints three bullets. They describe four separable behaviours:

Behaviour	Google’s wording
Incentives	“Offer incentives – such as payment, discounts, free goods and/or services - in exchange for posting any review or revision or removal of a negative review.”
Gating	“Discourage or prohibit negative reviews, or selectively solicit positive reviews from customers”
On-premises pressure	“merchants should not require or pressure users to leave ratings or write reviews while on the premises”
Specific content	“nor should they request that specific content be included”

Incentives. The bullet turns on the exchange, not on the rating. A discount offered for any review at all is inside it, and so is one offered for changing or deleting a review already posted. A review a competitor obtained that way is reportable, which is how to report a fake Google review, and what Google will delete is in removing a Google review.

Gating. The survey-first funnel: a product asks “How did we do today?”, routes happy answers to your Google review link and unhappy ones to a private complaints form. Google’s own words for it are “selectively solicit positive reviews from customers”.

On-premises pressure. The full sentence is one clause longer than the table row: “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.” That is the iPad by the exit with a staff member watching, which is why Google’s own in-store format is a printed QR code the customer can ignore.

Who you may ask

The other way a request goes wrong is who gets asked. Three prohibitions, these from the Fake Engagement section:

• “Content that has been posted from multiple accounts by or at the request of one person.”
• “Content that is based on a conflict of interest.” Google’s examples include current or former employment, contractual relationships, industry competitors and familial relationships.
• “Content exhibiting unusual volumes or patterns of review contributions that are indicative of efforts to manipulate a place’s rating”

Asking your staff, your family or your own second account is each of those in turn. Google names the management version too, prohibiting “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.” A review quota for the front desk is a violation in Google’s own words, and so is asking to be mentioned by name.

Two wordings of the same warning

Google prints the incentive warning at the top of both review-request pages, in two different vocabularies. The tips page says offering incentives “is considered fake & misleading content and is strictly prohibited”. The link and QR code page says the same practice “is considered fake engagement and is strictly prohibited”. Both sentences then link to a page titled “Business Profile restrictions for policy violations”, which we have not read.

The policy page repeats itself too. The identical merchant bullet about offering incentives is printed twice, in two adjacent sections: Fake Engagement, and Rating Manipulation, which opens “Rating Manipulation includes Incentivized or Biased Reviews, which are not allowed and will be removed from Maps.”

No numbers, because there are none

There is no response rate, open rate or conversion figure here, because none of the Google pages cited publishes one, and we will not restate a number from somewhere else as though it described your business. Google does publish annual, platform-wide counts of reviews it blocked or removed, quoted in what a fake review checker cannot do.

What Google says about asking well is qualitative: “Honest and balanced reviews can help potential customers decide. A mix of positive and negative feedback often feels more trustworthy.” That is an argument against gating, made by the platform that would enforce against it.

Why this product does not do any of it

mesej never messages anyone first, which is structural rather than a missing feature: a request sent to somebody who has not written to you is an originated message, and the system has no path that originates one. Stated exactly, because the version here said never at all until this pass: SendKind in src/lib/conversation/send_guard.ts carries a review_ask value, so the specification does allow asking inside a conversation the customer already opened. Never first is the guarantee.

What we are specified to build is the reply side, where Google moderates what you write and can reject it, as in what Google publishes when a reply is rejected and how to respond to a negative review. Specified is the accurate word: there is no Business Profile code in the repository, so that half is a design rather than a feature.

If a vendor in this category offers to filter who gets asked, run an incentive campaign, or script the wording a customer should use, the quoted rules above are the standard to hold it to.

The short version

Ask. Use the link or the QR code from your own profile, put it on a receipt, in a thank-you email, at the end of a chat, or on a printed sign. Attach nothing, ask everyone rather than the customers you expect to be happy, do not stand over anybody while they type, and do not tell them what to write.