Key Takeaway
Audit product pages and review components. Remove fake reviews, clearly disclose benefits such as cash, discounts, vouchers or free products near relevant genuine reviews, and align Review and AggregateRating markup with visible content. Stars in code, hidden disclosures and ratings aggregated from other sites do not meet the cited guidance.
What's the official change?
Google's July 24, 2026 update added guidance on fake and undisclosed incentivized reviews to improve transparency. It concerns review-snippet eligibility and applies to both visible content and structured data.
The guidance distinguishes reviews not based on genuine experience from reviews written for benefits without prominent disclosure. An incentive is not, under this clause alone, an automatic prohibition; disclosure also does not override other product, platform or legal requirements. As an implementation safeguard, disclose the relationship where the review is read, not only in a footer or separate campaign policy.
Existing requirements still apply: marked-up reviews and ratings must be visible, concern a specific item rather than a category or list, and not aggregate ratings from other websites. Google recommends review text and an author's name with ratings. Violations may lead to manual action, with reconsideration available after correction.
Implications for International Brands
Review collection and search presentation need one factual record. Trial, gift, coupon or loyalty campaigns may record incentives that frontend components omit. Share review and product IDs, market, language, experience-verification status, incentive type, disclosure and moderation state across growth, support, legal, development and data teams.
Do not combine ratings from Amazon, Google Business, Facebook, dealers or directories into the site's AggregateRating. External reviews may be linked or attributed separately where permitted; the site's rating should be traceable to eligible, visible and auditable review records.
Map reviews to the correct product or service. Do not indiscriminately combine categories, model families, market versions or distinct variants. Reconcile SKUs, GTINs, models and variants, with explicit scope and deduplication rules when the same eligible review set appears across pages.
Review transparency concerns eligible search presentation, not a promised AI-citation signal. The update does not establish that compliant reviews improve rankings, appear in AI Overviews or AI Mode, or receive a separate incentive-disclosure ranking benefit.
Zhihe Growth's Assessment
Zhihe Growth treats acceptance as an evidence-chain review: collection records, displayed reviews, markup and campaign terms must agree. Rich Results Test checks structure, not actual product experience, disclosure sufficiency or rating provenance.
Use a clear moderation policy: hold reviews pending verification rather than labeling every unverified review fake; reject demonstrably fabricated content; retain genuine incentivized reviews only with appropriate disclosure and compliance checks; and keep external-platform ratings out of the site's aggregate. Localized disclosures must remain visible on mobile, collapsed components and pagination states.
Markup must not expose an aggregate inconsistent with visible eligible reviews. After legitimate removal, correction, migration to a new SKU or domain, or a change in scope such as removal required by regional law, recalculate the aggregate rating, reviewCount and ratingCount, and synchronize visible pages and structured data. Otherwise, even genuine individual reviews can produce misleading totals and different facts for users and search systems. A refund or negative experience does not by itself invalidate a genuine review and must not be used to suppress unfavorable feedback.
Review publication and markup decision matrix
| Review status | Page treatment | Markup treatment | Acceptance boundary |
|---|---|---|---|
| Genuine experience, no incentive | Display review text, rating and author identifier | Use the applicable supported type and eligible aggregate | Meets this transparency condition; other guidelines still apply |
| Genuine experience with a prominently disclosed incentive | Explain the cash, discount, gift or other benefit near the review | Mark up only the review and aggregate actually visible | Not automatically excluded by this clause; retain evidence and check other rules |
| Genuine experience with an undisclosed incentive | Add clear, prominent disclosure before publication | Exclude from markup and aggregates until corrected | Does not meet the cited transparency guidance |
| Genuine experience not yet established | Hold for verification; reject fabricated reviews rather than relying on disclaimers | Exclude ineligible Review entries and recalculate AggregateRating | Unverified is not proven fake; genuine experience must be established before acceptance |
| Ratings from other websites | Link or attribute externally where permitted; do not present as the site's own rating | Do not combine into the site's AggregateRating | Avoid cross-site aggregation |
Implementation Checklist
- Export review records. Track review ID, product ID, SKU, market, language, author, rating, text, actual publication time, source channel and current URL.
- Retain authenticity evidence. Keep order, trial, service-delivery or other experience evidence securely, separate from public reviews. Disclose only necessary information.
- Record incentives. Record cash, discounts, vouchers, points, gifts, prize eligibility and other benefits, along with campaign, market and disclosure text.
- Check disclosure visibility. Verify clear, prominent disclosures near the review across desktop, mobile, pagination, collapsed states and translations.
- Remove fabrication and cross-site aggregates. Investigate unverified content and remove confirmed fake reviews. Do not copy third-party ratings into the site's AggregateRating.
- Identify the reviewed entity. Map each review to the actual product or service, checking model, variant, GTIN, market and canonical relationships.
- Synchronize content and markup. Keep visible ratings, counts, authors and review text consistent with Review, AggregateRating, ratingCount and reviewCount.
- Define moderation states. Track pending verification, organic reviews, disclosed incentives, rejection, deletion, appeals and corrections; recalculate aggregates when eligible membership changes.
- Validate templates. Test representative pages with JSON parsing, Rich Results Test and URL Inspection, and manually check mobile presentation before wider rollout.
- Monitor without guarantees. Record valid and invalid items and search performance in Search Console. Separate rich-result display from ordinary impressions; promise neither stars nor AI citations after correction.
Limitations
The original editorial log recorded a monitoring cutoff of July 28, 2026 at 23:58 Beijing time and reported no more suitable major official update in its preceding 72-hour window. That historical monitoring claim was not re-executed for this translation. The article uses Google's July 24 update as an evergreen topic, not as a newly published event inside that 72-hour window.
Meeting markup requirements does not guarantee rich results. Search Essentials, general structured-data guidance and type-specific rules still apply, including restrictions on self-serving Organization or LocalBusiness review stars. This decision matrix is Zhihe Growth's implementation guidance, not Google's approval tool or a substitute for market-specific legal and platform-policy review.
This update concerns Google Search review snippets, not citation rules for AI Overviews, AI Mode, Gemini or other systems. Genuine, transparent reviews provide better evidence but do not guarantee rankings, citation rates or conversions. Assess outcomes after publication and recrawling with a sufficient observation window and comparable measurements.
Official sources
- Google Search Central:Latest documentation updates Source for the July 24, 2026 update and its transparency purpose.
- Google Search Central:Review snippet structured data Source for authenticity, incentives, visibility, item scope, aggregation, self-serving reviews and monitoring limits.