Understanding DSA Article 27: The Transparency Mandate
Article 27 DSA obligates online platforms to be transparent about their recommender systems. This fundamental requirement has reshaped how platforms must communicate with users about algorithmic decision-making. But what exactly does this obligation entail, and how should platforms of all sizes approach implementation?
Article 27 of the DSA mandates all online platforms (not just VLOPs/VLOSEs) to disclose, in "plain and intelligible language," the main parameters of their recommender systems (including their relative importance) as well as any options for end-users to change them. This baseline requirement applies across the digital ecosystem—from large social media platforms to smaller niche services.
The Legal Framework: What Article 27 Requires
Understanding the specific obligations under Article 27 is essential for compliance. The regulation sets out a clear, three-pronged approach to recommender system transparency:
Main Parameters and Plain Language Disclosure
Providers of online platforms that use recommender systems shall set out in their terms and conditions, in plain and intelligible language, the main parameters used in their recommender systems, as well as any options for the recipients of the service to modify or influence those main parameters.
The plain language requirement is critical. This isn't an invitation to use technical jargon or bury information in lengthy legal documents. The purpose of Article 27 of the DSA is to enable recipients to understand how certain information is prioritised for them and how their online behaviour influences the recommendation of certain products, services, or content. Platforms must translate algorithmic concepts into language that ordinary users can comprehend.
Explaining Criteria and Relative Importance
The main parameters shall explain why certain information is suggested to the recipient of the service. They shall include, at least: (a) the criteria which are most significant in determining the information suggested to the recipient of the service; (b) the reasons for the relative importance of those parameters.
This requirement goes beyond simple disclosure. Platforms must explain not just what factors matter, but why they matter. For instance, rather than stating "engagement signals," a platform should explain that content receiving higher engagement may be ranked higher because it indicates user interest. Parameters might include watch history, time spent, geographical location, or user ratings.
User Modification and Control Functionality
Article 27 requires platforms to clearly explain the main and most significant parameters used in their recommender systems, and to allow users to directly and easily select or modify their preferred recommendation settings when multiple options are available. Where multiple recommender system options exist, providers of online platforms shall also make available a functionality that allows the recipient of the service to select and to modify at any time their preferred option.
This control mechanism must be practical. That functionality shall be directly accessible from the specific section of the online platform's online interface where the information is being prioritised. Users shouldn't need to navigate deep into settings—the option to change recommendation parameters should be immediately visible where recommendations appear.
Scope and Limitations: Understanding the Boundaries
While Article 27 represents a significant transparency mandate, it's important to understand its scope. The requirement for transparency within terms and conditions does not apply to providers of online search engines or VLOSEs; and does not require platforms to provide information on each and every parameter used or provide information on how individual parameters are weighted within such systems.
This distinction matters. Search engines operate under different rules, and platforms are not required to disclose granular technical weightings. The focus is on explaining main parameters in understandable terms, not providing a mathematical breakdown of algorithmic formulas.
Practical Implementation: A Roadmap for Platforms
Step 1: Audit Your Recommender Systems
Before drafting disclosures, conduct a comprehensive audit of all systems that suggest, prioritize, or determine the relative order of information. This includes:
- Content feeds and timelines
- Search result rankings
- Product recommendations
- Personalized advertising systems
- User suggestion algorithms
Step 2: Document Main Parameters and Their Rationale
For DSA compliance purposes, your recommender system documentation should cover: What inputs does the system receive (user signals, content metadata, business rules)? What does it output (ranked list, score, filtered subset)? Create an internal parameter registry for each system (inputs, weights, override rules, re-training cadence).
The parameter registry becomes your internal source of truth. For each identified parameter, document not just what it is, but why it matters to your ranking algorithm.
Step 3: Translate Technical Concepts into Plain Language
This is where many platforms struggle. For Article 27 disclosures, specificity and consistency is key. The Better Feeds guidelines describe how platforms should disclose specific input data and weights in ways that allow for baseline interpretation of recommender systems' main parameters.
Rather than writing "machine learning model optimizes for engagement," explain: "Our system recommends content that similar users have interacted with. Interactions include likes, shares, comments, and time spent. Content you've engaged with influences what we show you next."
Step 4: Design User-Facing Control Mechanisms
If your platform offers multiple recommender system options, users must easily switch between them. Common examples include:
- Chronological vs. algorithmic feeds
- Personalized vs. generic recommendations
- Interest-based vs. trending content
- Engagement-optimized vs. diversity-balanced rankings
The control interface should be intuitive, accessible, and respect user choices immediately upon selection.
Step 5: Update Terms and Conditions and Privacy Notices
Disclose: Update your Terms of Service or add a standalone "How our recommendation system works" page per Art. 27. Many platforms create dedicated pages explaining recommender systems rather than embedding everything in lengthy terms and conditions.
Align these disclosures with your GDPR privacy notices to avoid conflicting information about data usage and algorithmic decision-making.
Implementation Challenges and Variations
These disclosure requirements can be interpreted in multiple ways, however, and the first round of DSA audits reveals variation in how platforms defined these expectations in the absence of clear regulatory guidance. This variation highlights an important challenge: Article 27 leaves room for interpretation, and different regulators may have different expectations.
At present, the design features implemented by very large social media platforms for the explanation and user control of their recommender systems seem to represent a mere case of transparency washing. While the DSA can still drive a historical shift in the experience of social media users, a set of guidelines, a code of conduct, or even a delegated act on how to ensure transparency and user control of recommender systems in online platforms would be essential to enable a meaningful implementation of Articles 27 and 38.
Platforms should therefore go beyond minimum compliance. Meaningful transparency means actually helping users understand—and control—their algorithmic experience.
Connecting Article 27 to Broader DSA Obligations
Article 27 doesn't exist in isolation. It connects to several other compliance requirements that platforms should coordinate:
Independent Audits (Article 37): The DSA mandates independent audits to assess the VLOPs/VLOSEs' DSA compliance. Auditors are tasked with evaluating the adequacy and effectiveness of the measures implemented to address identified risks, as well as the transparency of algorithmic processes. Your Article 27 disclosures will be scrutinized during audits. See our guide to independent audits for VLOPs and VLOSEs for detailed preparation strategies.
Researcher Data Access (Article 40): Article 40 DSA establishes a framework for data access, enabling regulators and vetted researchers to obtain data for the purpose of assessing compliance with the DSA from VLOPs/VLOSEs. Transparency about your recommender systems may be examined by researchers investigating algorithmic fairness and systemic risks. Our Article 40 implementation guide covers data access obligations in detail.
Systemic Risk Assessment (Articles 34-35): VLOPs and VLOSEs must assess and mitigate systemic risks, including those posed by recommender systems. Consistent disclosures could enable more effective risk assessment and mitigation under Articles 34 and 35. Our systemic risk assessment guide explains how recommender system transparency fits into this broader framework.
Enforcement and Penalties: Non-compliance with Article 27 carries serious consequences. If platforms were to fail in providing information requested by regulators, a formal initiation of proceedings could be started (Article 66 DSA). Eventually, the Commission can impose fines for incorrect, incomplete, or misleading responses to these requests for information (Article 74(2) DSA). For real-world examples of enforcement actions, consult our 2026 enforcement case studies.
Platform Size Matters: Differentiated Approaches
While Article 27 applies to all online platforms, implementation strategies can vary by size:
Small and Medium Platforms
If your platform has limited resources, focus on clarity and honesty. Document your recommender systems in plain language. If you use a standard ranking algorithm (such as "most recent" or "most relevant"), explain it simply. If you use machine learning, explain what inputs drive ranking without claiming technical sophistication you don't possess.
Large Platforms (VLOPs)
VLOPs face heightened scrutiny. Your disclosures must be comprehensive, and your control mechanisms must be genuinely functional. Article 38 mandates that VLOPs/VLOSEs must offer at least one version of their recommender system that does not rely on "profiling" under the General Data Protection Regulation (GDPR). This non-profiling option must be clearly disclosed and easily accessible.
Best Practices for Meaningful Implementation
Beyond legal compliance, platforms achieving genuine transparency follow these practices:
- Use layered explanations: Provide a simple summary for most users, with optional deeper technical details for interested audiences.
- Be specific about data sources: Explain what specific user actions (clicks, time spent, shares) influence recommendations.
- Acknowledge limitations: Users benefit from knowing when systems have incomplete information or when recommendations reflect business objectives alongside relevance.
- Test user comprehension: Don't assume your explanations are clear—gather user feedback on whether they actually understand your recommender systems.
- Update regularly: As algorithms change, update your disclosures. Stale information undermines transparency.
- Make control intuitive: If users can't easily modify their recommendation preferences, you're not really giving them control.
Looking Forward: Consistency and Standardization
The DSA ecosystem is still evolving. For Article 27 disclosures, specificity and consistency is key. Consistent disclosures would allow independent experts, users, and the European Commission to examine and compare how recommender systems are optimized across platforms.
As enforcement matures and guidance clarifies, expect regulators to push for consistency across platforms. Platforms investing in robust Article 27 compliance now will be best positioned as the regulatory landscape becomes more precise.
For the latest DSA developments and compliance guidance, visit our blog.
Conclusion
DSA Article 27 represents a fundamental shift toward algorithmic transparency. Rather than viewing it as a compliance burden, platforms can see it as an opportunity to build user trust. Users increasingly demand to understand how algorithms shape their online experience. Clear, honest, and comprehensive disclosures about recommender system parameters and user control options serve both regulatory compliance and user satisfaction.
The requirement to explain "main parameters" in plain language isn't just legal theater—it's a genuine effort to demystify algorithmic recommendation. Platforms that embrace this transparency, rather than treating it as a minimum requirement, will emerge as trusted players in the digital services ecosystem.