Google added a new review snippet guideline on 24 July 2026: pages and structured data should not include fake reviews or incentivized reviews when the incentive is not clearly and prominently disclosed. Google said the change was intended to improve review transparency.
The wording matters. Google did not announce a new ranking factor, a new review-specific manual action, or a blanket ban on every incentivized review. The immediate SEO issue is review rich result eligibility, and the July change sits alongside existing rules about genuine review content, visible markup, review sourcing, and self-serving reviews.
For the durable basics of choosing and validating structured data, see MOCOBIN’s Schema markup guide. This analysis is narrower: it explains what Google’s July review policy clarification changes for review collection, publishing, and structured data workflows.
What Google Added on 24 July 2026
Google’s Search documentation changelog records a new review snippet guideline for fake and undisclosed incentivized reviews. The current review snippet documentation gives two examples:
- reviews that are not based on a genuine experience with the product or service;
- reviews written in exchange for a benefit, such as money, discounts, vouchers, or free products, when that incentive is not clearly and prominently disclosed.
The second example is easy to overread. The new wording targets undisclosed incentivized reviews. It does not say that disclosure automatically makes every incentivized review eligible, nor does it define one universal disclosure label or placement. The review still has to satisfy the rest of Google’s review snippet and structured data requirements.
That distinction changes the first question in an audit. Instead of asking only whether the JSON-LD validates, identify how the review was obtained and whether the page tells users about any benefit connected to it.
Do Not Confuse the New Rule With Existing Review Restrictions
Several important review rules were already part of Google’s documentation and remain separate from the 24 July addition. Combining them into one ‘new policy’ makes diagnosis harder.
| Rule | Status | What it means operationally |
|---|---|---|
| Fake or undisclosed incentivized reviews should not appear on the page or in structured data. | Added to the review snippet guidelines on 24 July 2026. | Review collection and disclosure now need to be checked before the markup layer is approved. |
| Marked-up review content must be readily available to users on the page. | Existing review snippet requirement. | The visible review, rating, and structured data must describe the same content. |
| Do not aggregate reviews or ratings from other websites. | Existing review snippet requirement. | Importing an external rating does not make it eligible for markup simply because a widget can display it. |
Self-serving reviews for LocalBusiness or Organization are ineligible for the star review feature when the reviewed entity controls the reviews about itself. |
Existing restriction, not the July 2026 addition. | A genuine, disclosed testimonial on a business’s own site can still be ineligible for this particular rich result use. |
This separation is important for local businesses. Google’s current documentation explicitly includes reviews placed on the organisation’s own site through an embedded third-party widget in its self-serving example. A trusted widget does not override the eligibility rule.
Which Review Workflows Need Attention First
The policy change matters most where a business has more than one path for collecting reviews. The page may look simple while the review data has passed through email campaigns, loyalty systems, sampling programmes, marketplaces, moderation tools, and a schema plugin.
E-commerce sampling, discounts, and loyalty campaigns
If a customer received money, a discount, a voucher, a free product, or another benefit in exchange for writing a review, the July wording makes the disclosure question explicit. The incentive should be clearly and prominently disclosed, and the review must still represent a genuine experience.
The operational risk is inconsistent classification. A campaign system may know that a review came from a free-product programme while the storefront receives only the rating, text, and reviewer name. If the incentive flag disappears before publication, the final page and structured data can no longer represent the collection context accurately.
Editorial scores and customer ratings
Publishers should keep editorial evaluations and user ratings conceptually separate. An editor’s score is not automatically the same data as an aggregate rating from customers. If the review itself was produced in exchange for a benefit, the new incentive-disclosure wording still needs to be considered, but the July update should not be stretched into a general rule about every commercial relationship on the page.
Local business and organisation reviews
For LocalBusiness and Organization, two checks are required. First, is the review genuine and, where relevant, is an incentive disclosed? Second, is the page eligible for Google’s star review feature at all?
Those questions can produce different answers. A testimonial may be useful to visitors and transparently collected while remaining ineligible for self-serving review markup on the reviewed organisation’s own site.
Audit the Review Pipeline, Not Just the Plugin Screen
Third-party tools can generate Review or AggregateRating markup, but the final rendered page is what matters. The safer audit sequence starts with the data source and ends with the rendered output.
- Inventory review-bearing templates. Identify product, service, location, editorial, and other pages that output ratings or review structured data.
- Classify the source. Record whether reviews come from direct submissions, verified purchases, sampling campaigns, loyalty incentives, editorial authors, or imported systems.
- Check authenticity and incentives. Confirm that the review reflects a genuine experience and that any relevant benefit is clearly and prominently disclosed.
- Check feature eligibility. Confirm that the reviewed item is supported, that the review refers to a specific item, that the site is not aggregating ratings from another website, and that local business or organisation markup does not fall into the self-serving restriction.
- Compare visible content with rendered markup. The rating, review text, reviewed item, author information, and aggregate values should not conflict with what users see.
- Test representative live URLs. Use Google’s Rich Results Test and URL Inspection where appropriate. One passing example is not enough when several templates or data sources are involved.
- Assign ownership. Decide who controls review collection rules, disclosures, widget configuration, structured data, and post-release checks.
The order matters. Fixing JSON-LD before establishing where the review came from can produce technically neat markup around a policy problem that started much earlier in the workflow.
If Review Stars Disappear, Do Not Assume a Penalty
Google does not guarantee that valid structured data will produce a rich result. Its general structured data guidelines also distinguish rich result eligibility from web search ranking.
That gives review teams three different situations to diagnose:
- No star result: this alone does not prove a policy violation or manual action. Google may simply not show the feature.
- Structured data error or eligibility problem: inspect the feature requirements, rendered page, Rich Results Test, and relevant Search Console reporting.
- Manual action: treat it as confirmed only when the Manual Actions report shows one. Google says a structured data manual action removes rich result eligibility but does not, by itself, change how the page ranks in Google web search.
A visibility decline should therefore be investigated against the date and scope of the change. Check which templates lost review eligibility, whether the review source or widget changed, and whether the affected pages share the same collection method. Removing every review because stars disappeared is usually the wrong first move.
What Still Needs Monitoring
The confirmed change is narrow: Google added the fake and undisclosed incentivized review guideline on 24 July 2026. The 24 July changelog entry does not describe the update as a ranking signal and does not announce a new review-specific manual action category.
Site owners should watch for three things: further changes to the review snippet documentation, updates from review platforms that alter their markup or incentive labelling, and Search Console evidence on the site’s own affected templates. A platform update may repair a field without fixing the collection process that produced the data.
The practical checkpoint is simple. Before review markup is approved, the team should be able to answer where the review came from, whether a benefit influenced its creation, what the user can see, and whether that page is eligible for the Google feature being targeted. If one of those answers is unclear, validation should wait.
