Google added a new guideline to its review snippet structured data documentation on July 24, 2026, prohibiting fake reviews and incentivized reviews that fail to disclose the incentive. The rule sits inside the guidelines section that governs eligibility for star ratings in search results, where violations expose a site to manual action.

The change is short. Its placement is not.

According to Google's Search Central changelog, the company on July 24, 2026 added a new guideline to the review snippet documentation about fake and undisclosed incentivized reviews, with the stated purpose of improving user review transparency. The review snippet documentation page carries a last updated stamp of 2026-07-24 UTC, matching the changelog entry.

What the new guideline says

The added text appears in the guidelines section of the review snippet page, the part of the document that determines whether marked-up content qualifies for a rich result at all. According to the documentation, sites should not include fake or undisclosed incentivized reviews on the page or in the structured data markup.

The documentation then supplies two examples of what falls inside the prohibition. The first covers "Reviews that aren't based on a genuine experience of a product or service". The second covers "Reviews written in exchange for a benefit (such as money, discounts, vouchers, or free products) that don't clearly and prominently disclose the incentivization".

Two elements of that phrasing carry weight. Incentivized reviews are not banned outright. What is banned is incentivization that goes undisclosed, or that is disclosed without clarity and prominence. A site running a product seeding programme, a voucher-for-feedback campaign, or a sampling scheme is not automatically outside the guideline, provided the arrangement is visible to the reader. The second element is the reach of the phrase "on your page or in your structured data markup", which extends the rule past the JSON-LD block and onto the visible page itself.

Why the guidelines section matters

The review snippet page opens its guidelines section with a warning. According to the documentation, if a site violates one or more of these guidelines, Google may take manual action against it, and once the problem has been remedied the site can be submitted for reconsideration.

The consequences of a structured data manual action are set out in the troubleshooting section of the same page. According to the documentation, when a page receives a structured data manual action, the structured data on that page will be ignored, although the page can still appear in Google Search results. Rankings are not the exposure. Rich result eligibility is.

For sites whose organic click-through performance depends on star ratings appearing beside the blue link, that distinction is the whole story. A review snippet is a visual differentiator in a result set that increasingly compresses organic listings beneath AI-generated answers and shopping modules. Losing the stars does not remove a page from the index. It removes the reason a user picks that page over the one above it.

The existing guideline set

The new prohibition joins a list that had already grown restrictive. According to the documentation, aggregate evaluations of an item by many people must be marked up with schema.org/AggregateRating, reviews must refer clearly to a specific product or service rather than a category or a list of items, and marked-up review content must be readily available to users from the marked-up page. Sites that include multiple individual reviews are also expected to include an aggregate rating of those reviews, and the documentation states plainly that reviews or ratings from other websites must not be aggregated.

A separate cluster of rules applies to local businesses and organizations. According to the documentation, if the entity being reviewed controls the reviews about itself, pages using LocalBusiness or any other Organization structured data are ineligible for the star review feature. The documentation gives the example of a review about an entity placed on that entity's own website, either directly in structured data or through an embedded third-party widget. It further states that ratings must be sourced directly from users, and that sites must not rely on human editors to create, curate, or compile ratings information for local businesses.

The self-serving review restriction has been in place for years. The July 24 addition operates on a different axis. Where the older rule asks who controls the review, the new one asks what the reviewer received in exchange for writing it.

Technical scope of the affected markup

The guideline applies across a wide surface. According to the documentation, ratings can be supplied for Book, Course list, Event, Local business, Movie, Product, Recipe, and Software App features. Google additionally supports reviews for CreativeWorkSeason, CreativeWorkSeries, Episode, Game, MediaObject, MusicPlaylist, MusicRecording, and Organization schema types, along with their subtypes.

The property requirements underneath have not changed. For Review markup, the documentation lists author, itemReviewed, itemReviewed.name, reviewRating, and reviewRating.ratingValue as required properties, with datePublished, bestRating, and worstRating recommended. The author field must be shorter than 100 characters, and the documentation is explicit that a promotional string is not an acceptable reviewer name, citing "50% off until Saturday" as an example of what will not qualify.

For AggregateRating markup, the required properties are itemReviewed, itemReviewed.name, ratingValue, and at least one of ratingCount or reviewCount. The default rating scale is a five-point scale unless bestRating and worstRating are declared, and decimal values must use a dot rather than a comma, with Microdata and RDFa implementations able to use content attributes to override the visible display.

Those specifications matter to the new guideline in a practical sense. The disclosure requirement attaches to review content, and review content is what the author, reviewRating, and reviewBody fields carry into the index. A compliance review that stops at the schema validator will not surface a disclosure problem, because the Rich Results Test checks syntax rather than provenance. The documentation itself acknowledges this limit, noting in its troubleshooting section that the issue may not be a syntax issue and that the tool will not identify it.

Eighteen months of tightening

Google last altered the same guidelines section in January 2025, when it added a recommendation that ratings be accompanied by a review comment and the author's name. That change was framed as a recommendation rather than a requirement, and the current documentation retains it in that form: the company recommends only accepting ratings accompanied by a review comment and author's name, while noting the approach is not required.

The July 24 addition is worded differently. It is a prohibition, not a recommendation, and it sits above the warning about manual action rather than below it.

The July documentation cadence

The review snippet guideline is the fifth Search Central documentation change published in July 2026, and the pattern across the month is consistent. On July 1, Google simplified its AMP documentation by removing references to the AMP viewer, AMP Cache, and signed exchange, alongside a routing change sending Search users directly to publisher-hosted AMP pages. On July 7, the company added the CategoryCode type to Merchant listing structured data and published a new section on sale duration covering the validFrom, validThrough, and priceValidUntil properties. On July 10, it clarified that canonical re-evaluation can take up to two weeks after a publisher fixes the underlying content. On July 14, it closed the package tracking early adopters programme to new partners while leaving existing participants in place.

Read together, the month shows commercial structured data being extended while reference-oriented features are frozen or withdrawn. That trajectory has been visible for more than a year. Google deprecated seven structured data types in a single action in June 2025, including Book Actions, Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement, and Vehicle Listing. It removed practice problem structured data support beginning in January 2026. FAQ rich results stopped appearing in Google Search on May 7, 2026, with documentation removed the following month.

Review and AggregateRating markup survived that contraction, and adoption data explains why. Schema.org and Google published domain-level usage figures in June 2026 showing both Review and AggregateRating in the one million to ten million domain bucket, alongside Product, Offer, and LocalBusiness. Review markup is not a niche feature Google can quietly retire. It is infrastructure, which makes guideline enforcement the only available lever.

Enforcement already running elsewhere

The new guideline lands into an enforcement environment that has been escalating on adjacent surfaces for over a year. Google Maps deployed Gemini models in April 2026 to catch manipulated place edits and to block review extortion scams before publication, published alongside 2025 Trust and Safety Report figures recording more than 292 million policy-violating reviews blocked or removed against more than one billion published, plus 79 million blocked edits and over 13 million fake Business Profiles removed.

One day later, Google added clauses to the Maps user-generated content policy banning quota-based review solicitation and reviews that mention staff by name. In June 2026, the platform began surveying users about whether businesses had offered anything in exchange for a review. The underlying Maps position on incentives was already unambiguous: help documentation formalized on December 31, 2025 states that offering incentives, such as free or discounted goods or services, to customers in exchange for reviews is considered fake engagement and is strictly prohibited.

Until July 24, that position governed Business Profile content. It did not govern the first-party review systems publishers and retailers run on their own domains. The new guideline closes that gap on the structured data side, though with a softer rule: disclosure, rather than outright prohibition.

Related documentation work has run in parallel. Google clarified on May 15, 2026 that its spam policies apply to generative AI responses in Search, and on June 5, 2026 published guidance on evaluating third-party SEO tools, services, and advice.

Regulatory alignment

The wording of the new guideline tracks existing law in Google's largest advertising market. The FTC finalized a rule on consumer reviews and testimonials on August 14, 2024 that prohibits creating, selling, purchasing, or distributing fake consumer reviews and testimonials, and allows the agency to pursue civil penalties against knowing violators. The agency estimated the rule could save consumers between 6.64 billion and 17.52 billion dollars annually.

In Europe, the picture is more contested. German businesses have used Digital Services Act notification mechanisms to remove unfavourable reviews at scale, and Google Maps subsequently began displaying removed review counts on German business profiles. The proposed Digital Fairness Act would add another layer, and the advertising industry has already objected to the scope of the consultation, arguing that existing frameworks cover the same ground.

A platform guideline is not law, and Google is not a regulator. What the July 24 addition does is align rich result eligibility with disclosure standards that consumer protection authorities were already enforcing through different instruments, which narrows the gap between what is legally required and what is technically rewarded.

What changes for practitioners

Sites operating review programmes with any incentive component now carry two separate compliance obligations rather than one. The legal obligation, where applicable, requires disclosure of material connections. The platform obligation requires that disclosure to be clear and prominent on the page and reflected in the structured data, with rich result eligibility as the stake.

Ecommerce operations running product sampling, subscription boxes with feedback requests, or affiliate review programmes sit closest to the line. So do software review aggregators, course platforms, and any site whose star ratings were assembled from user feedback solicited with a benefit attached.

The enforcement mechanism remains manual action, which means a human reviewer and a Search Console notification rather than an algorithmic adjustment. That has historically meant slower and more selective application than automated systems allow. The precedent of the site reputation abuse policy is instructive: written first, enforced manually later, with algorithmic detection developing over the following years. Whether the review snippet guideline follows the same curve is not yet observable.

Timeline

Summary

Who: Google published the change through Search Central documentation. It affects website owners, retailers, ecommerce operators, review aggregators, software directories, course platforms, and any publisher using Review or AggregateRating structured data to display star ratings in search results.

What: A new guideline in the review snippet documentation prohibits fake reviews and incentivized reviews that do not clearly and prominently disclose the incentivization, covering both visible page content and structured data markup. Violation exposes a site to manual action, which causes the structured data on the affected page to be ignored while leaving the page in search results.

When: The guideline was added on July 24, 2026, and the review snippet documentation page carries a last updated stamp of the same date. It is the fifth Search Central documentation change published in July 2026, following entries on July 1, July 7, July 10, and July 14.

Where: The change applies to Google Search globally, across every feature that supports review snippets, including Book, Course list, Event, Local business, Movie, Product, Recipe, and Software App, along with supported schema types such as Game, MediaObject, MusicRecording, and Organization.

Why: Google stated the purpose as improving user review transparency. The addition extends to first-party review systems a disclosure standard the company had already applied to Google Business Profile content, and aligns rich result eligibility with consumer protection rules such as the FTC final rule on consumer reviews and testimonials.