Yelp’s Trust & Safety report for 2025 (which is not EU-specific) can be found here.
How Yelp safeguards content integrity
Yelp’s mission is to connect consumers with great local businesses by giving them access to reliable and useful information.Yelp’s automated recommendation software
Yelp’s automated recommendation software evaluates all reviews on our platform using hundreds of signals related to quality, reliability, and user activity on Yelp. The system is entirely automated, applying the same objective rules to every business, and no Yelp employee, business owner, or reviewer can influence or override the software’s decisions for any individual business—this approach is deliberate to avoid conflicts of interest.The reviews recommended for any business can change over time, and may move between recommended and not recommended as the software learns more about the reviewer and the business. Reviews identified as less reliable or useful are displayed in a “not currently recommended” section and do not factor into a business’s overall star rating.
In 2025, we enhanced the software to better identify and not recommend reviews lacking sufficient details about the consumer experience or containing potential conflicts of interest—such as those written by individuals with undisclosed ties to the business or who received discounts or other incentives in exchange for their review. This helps us surface the most useful and reliable content, maintain the integrity of our platform, and protect both consumers and businesses from misleading information.
For more information about Yelp’s automated recommendation software, visit our website or watch this video for a brief overview.
Prioritizing Trust & Safety by mitigating AI-generated reviews
As AI becomes more accessible, it’s not only transforming how people engage online but also how bad actors attempt to manipulate platforms. Our longstanding content policies require all reviews and photos to reflect genuine, firsthand experiences. At Yelp, AI-generated reviews and photos submitted by users violate our content policies, and reviewers are prohibited from using third-party AI tools to write their reviews. While using tools to correct spelling and grammar are permitted, reviews that appear to be drafted or revised using third-party AI tools are not allowed, as content must be authentically created by the reviewer to keep Yelp helpful and reliable. Instead, to make review writing easier, we provide helpful features like interactive review topics that remind reviewers of helpful points to address—like “service,” “value,” or “care quality.” As reviewers address each topic, it’s checked off, helping capture more meaningful details in their reviews.With the rise in consumer adoption of AI tools, Yelp continues to significantly invest in methods to better detect and mitigate AI-generated reviews, including rolling out several new AI detection technologies. In 2025, Yelp’s automated systems identified and unrecommended nearly half a million reviews that exhibited characteristics of AI-generated content.* Our systems also prevented scams that disrupted other platforms—most notably, the review extortion incidents that impacted businesses on Google in major U.S. metros like Las Vegas, Philadelphia, and New York in 2025. In nearly all cases, businesses on Yelp were not affected, and in the rare instances where similar attempts surfaced, our systems thwarted suspicious activity before it could appear on business pages.
Beyond our automated systems, our User Operations team proactively audited and investigated suspicious business pages in 2025, uncovering review exchange rings and suspicious activity that led to a 49% increase in related account closures from 2024.† Meanwhile, media coverage and social platforms drove a surge in abnormal review activity, resulting in a 58% increase in Media Attention Alerts placed on business pages.‡ Our team responded by temporarily disabling the ability to publish content on affected pages and removing over 80,000 reviews that did not reflect firsthand consumer experiences.‡
As technology and tactics evolve, so does our approach. Our efforts to detect new deceptive tactics led to more than twice as many account closures compared to 2024, much of which was tied to large-scale airline phone support scams.*1 By combining advanced detection methods with hands-on human moderation and our robust content policies, Yelp continues to put user trust first by prioritizing authentic content—even as other platforms deprioritize trust and safety.
Yelp’s Content Moderation Governance Structure
Yelp’s content moderation governance structure is built on a collaborative and multi-tiered framework that addresses the accuracy, consistency, and reliability of content moderation decisions. At the core of this structure is Yelp’s User Operations team that works tirelessly to uphold the integrity of content on Yelp.Yelp’s User Operations Team
Yelp’s User Operations team is responsible for maintaining the reliability and integrity of content on the platform. The team’s responsibilities include protecting data quality, investigating potential fraud, and moderating content. They actively identify and remove content that violates Yelp’s Terms of Service and Content Guidelines, as well as detect, address, and remove bad actors from misleading Yelp consumers.The team proactively investigates and responds to consumer and business owner reports of suspicious behaviors or alleged illegal or harmful content, removing policy-violating content and closing user accounts that breach Yelp’s Terms of Service.
The team is structured for timely, accurate, and consistent moderation and is equipped to quickly address emerging trends and challenges.
The User Operations team collaborates closely with other teams such as engineering, legal, customer support, and data analytics to take a holistic approach to content moderation. This cross-functional effort helps keep Yelp a trusted platform for both users and business owners. More information about how Yelp’s User Operations team moderates content can be found here.
Maintaining accurate information on Yelp business pages
203 business pages removed in the EU in 20252 for being associated with lead generators, violating Yelp’s policies.Lead generators sometimes deceptively attempt to create fake business pages to generate leads from unsuspecting consumers and auction them to actual service providers. Typically posing as actual service providers, these lead generators often engage in undisclosed charges, provide inaccurate information about their identity and/or services, and impersonate reputable companies. Targeting vulnerable consumers needing emergency services, such as locksmiths, garage door repairs, or movers, these lead generators often exploit situations requiring home access. Our User Operations team regularly detects and blocks these sorts of bad actors on Yelp.
Review Removal and Account Closure
Approximately 307,000 reviews were contributed to EU business pages on Yelp in 2025.3- Recommended reviews: 31.6%4
- Not recommended reviews: 4.1%
- Reviews removed by reviewers themselves: 1%
- Reviews removed by Yelp: 63.3%
Distribution of reported reviews removed by reason in 20255
Approximately 1022 reviews reported by our community of users and business owners were removed from EU business pages by Yelp’s User Operations team.
Main Removal Reason |
% |
|
Represents an extraordinary experience This includes reviews that stem from a business gaining public attention
| 31% |
|
Contains False Information
While Yelp generally doesn't take takes in factual disputes and typically permits Yelpers to stand behind their reviews, Yelp will periodically remove reviews for containing false or misleading information
| 27% |
|
Other
This includes reviews that lack relevance (such as promotional content), spam or other matters that don’t address the core of the consumer experience | 23% |
|
Not a firsthand consumer experience
Yelp requires that reviews sufficiently reflect a firsthand consumer experience
| 9% |
|
Is for the wrong business
| 5% |
|
Posted by someone affiliated with the business, a competitor, or an ex-employee
| 3% |
|
Contains threats, lewdness, or hate speech
| 1% |
|
Violates Yelp's privacy guidelines
| 0% |
|
Duplicating reviews for multiple locations
This includes when a user posts the same content across multiple business pages
| 0% |
In 2025, we continued to enhance Yelp’s robust detection systems. The closure of these accounts in the EU for Terms of Service violations also resulted in the removal of more than 12,364 reviews.7
Approximately 141,600+ photos were removed in the EU as a result of account closures for Terms of Service violations in 2025.8
Tens of millions of photos are uploaded to Yelp annually,9 offering visual details that help inform people’s spending decisions.
Uncovering and reporting deceptive review groups on other platforms
Yelp actively investigates deceptive review practices both on our platform and third-party sites like Instagram, Facebook, X, LinkedIn, Reddit, TikTok, and Craigslist. Our User Operations team monitors and infiltrates online groups on these platforms where people may attempt to trade or pay for reviews, reporting any groups, posts, or individuals involved. We also flag instances of review selling or removal services originating from other sites.We have successfully worked with these platforms to address such behaviors and, in some cases, even improved their ability to prevent future occurrences. However, the final decision on addressing Yelp’s reports rests with each respective platform. Collaborating with these sites, as well as sharing leads on deceptive review conduct with the Federal Trade Commission (FTC) (a United States agency tasked with consumer protection) and other regulators, is essential to better protect consumers from misleading content. In 2025, we saw a 62% increase in action taken compared to 2024, underscoring Yelp’s proactive reporting and growing cooperation among platforms to combat deceptive review practices.*
1,020+ reports made by Yelp to other online platforms to warn them of content from 1,100+ suspicious groups, posts, or individuals we found on their sites, a 45% increase in suspicious content from 2024.*
60% of Yelp’s reports resulted in action being taken by online platforms to remove deceptive content in 2025.*
We have successfully worked with these other platforms to address deceptive behaviors and, in some cases, even improved their ability to prevent future occurrences. Partnering with these platforms is essential to better protect consumers from misleading content. However, final moderation decisions around Yelp-flagged content on other platforms lies with the respective platform.
For years, we have provided the FTC and other regulators with leads on deceptive review conduct, as well as submitted public comments to help strengthen the FTC's final rule, announced in 2024, banning fake reviews and testimonials.
Reigning in reputation management companies
Yelp strictly prohibits review gating and other review suppression tactics that disproportionately promote positive reviews, while diverting criticisms to private channels. These behaviors can result in misleading ratings and reviews about a business. We routinely investigate reputation management companies that offer these services, and if warranted, we take action against them, including reporting them to consumer protection agencies like the FTC. At times, Yelp also pursues legal action to disrupt these harmful markets.Yelp supports the FTC's enforcement of its rule on consumer reviews and testimonials, which prohibits businesses from suppressing or manipulating reviews. As the FTC continues to crack down on review manipulation, Yelp remains a proactive partner in helping identify and report bad actors that undermine trust in online reviews.
Yelp’s automated systems
To keep Yelp useful, relevant, and safe, we’ve developed automated systems to support human moderation in identifying promotional spam, inappropriate photos, and other content that violates Yelp’s Content Guidelines.Yelp uses machine learning as part of its protective measures to better detect reviews that may contain the most egregious instances of hate speech, vulgar language, or threats before they are ever published to Yelp. This system flags potentially inappropriate reviews to our moderators, and the moderator’s decision regarding the flagged content is then used to further train Yelp’s algorithms to more accurately flag potentially problematic content going forward. Since determining whether content is inappropriate can be context specific, we also continue to rely on reports from our community of users.
Yelp also has AI-based systems that are used to detect photos that may violate our Content Guidelines, including by proactively removing suspected inappropriate video content. These systems flag suspicious visual content for our moderators to review before it is published on Yelp.
Yelp constantly evaluates and monitors the performance of these automated systems to improve their accuracy. As a further safeguard, users are also able to appeal Yelp’s content and account moderation decisions as described here.
Member State Orders
Yelp did not receive any Member State Orders during the relevant period.Article 16 Notices
Information about Article 16 Notices can be found here.Content Moderation by Own Initiative (T&C)
Information about Content Moderation by Own Initiative (T&C) can be found here.Content Moderation by Own Initiative (Illegal Content)
Yelp’s content moderation systems primarily identify and remove content based on violations of its Terms of Service and Content Guidelines. Accordingly, metrics regarding Content Moderation by Own Initiative are reflected under the “T&C” classification rather than the “Illegal Content” classification.Appeals and Recidivism
Information about Appeals and Recidivism can be found here.Automated Means
Information about Automated Means can be found here.Machine Readable Version of Quantitative Report
A machine readable version of the quantitative element of Yelp’s transparency report can be found here.-------------------------------------------------------
* Data is sourced globally.
† Data is sourced from the U.S. and Canada.
‡ Data is sourced from the U.S. only.
[1] These account closures stem from both user reports and proactive investigations. When an account is closed due to suspected deceptive or abusive behaviors, the associated reviews are removed as well.
[2]This number is for the period of January 1 2025 through December 31, 2025.
[3] This number is for the period of January 1, 2025 through December 31, 2025, has been rounded and includes reviews that are currently recommended, not recommended, or that have been removed.
Much of this review removal and account closure was tied to the large-scale airline phone support scams referenced above.
[4] More information about Yelp’s Recommendation Software can be found here.
[5] This number is for the period of January 1, 2025 through December 31, 2025.
[6] Data is sourced from the EU and is for the period of January 1, 2025 through December 31, 2025.
[7] Data is sourced from the EU and is for the period of January 1, 2025 through December 31, 2025. These account closures stem from both user reports and proactive investigations. When an account is closed due to suspected deceptive or abusive behaviors, the associated reviews and photos are removed as well.
[8] This number is for the period of January 1, 2025 through December 31, 2025. Of these 141,600+ photos, 136,666+ were spam.
[9] Data is sourced globally for the period of January 1, 2025 through December 31, 2025.
* Data is sourced globally.
* Data is sourced globally.
* Data is sourced globally.