The platform offers an extensive suite of open-source SDKs for web, mobile, and backend environments, deeply integrated with feature flags and session recording.
As an engineering-focused platform, it provides a comprehensive library of open-source SDKs covering JavaScript, iOS, Android, React Native, Python, Node.js, and many others. Unlike pure measurement SDKs, these libraries are multi-functional; a single implementation natively handles event tracking, session recording, and the evaluation of feature flags without requiring separate scripts. This unified architecture significantly reduces payload size and simplifies the development workflow. Furthermore, because the platform is open-source, developers can actively inspect the SDK code, ensuring total transparency regarding how user data is collected, batched, and transmitted from the client to the server.
The platform uses a flexible event-based schema that uniquely combines automatic frontend capture (autocapture) with precise custom instrumentation.
Unlike traditional product analytics tools that strictly require manual tracking for every single action, this platform utilizes a hybrid tracking model. By default, its JavaScript snippet automatically captures all frontend interactions (clicks, pageviews, inputs) via 'autocapture', instantly populating the dashboard with data. Simultaneously, developers can define highly specific custom events and properties via code for critical backend actions (e.g., Payment Processed) or complex state changes. This gives teams the immediate visibility of an autocapture tool combined with the rigorous data quality of a deliberately instrumented custom schema. However, relying too heavily on autocapture without naming conventions can lead to cluttered, difficult-to-read reports.
Dashboards are highly customizable, collaborative workspaces that support SQL queries, Markdown text, and embedded session recordings.
The dashboard environment is designed to be a highly interactive workspace rather than a static reporting view. Analysts can pin various insights—including complex funnels, retention charts, or direct SQL queries—onto customized boards tailored for specific product squads. A major differentiator is the ability to interleave Markdown text blocks to explain the data context and directly embed relevant session recordings next to the charts. If a chart shows a sudden spike in errors, team members can watch the associated screen recordings immediately on the dashboard without switching contexts. This deep integration between quantitative metrics and qualitative visual evidence makes the dashboards exceptionally powerful for product diagnostics.
Users can build complex conversion funnels with advanced features like strict step-ordering, time-to-convert metrics, and direct links to session recordings.
Funnel analysis is a core feature, offering deep flexibility for analyzing multi-step user journeys. Analysts can define funnels using a mix of pageviews, custom events, or autocaptured frontend clicks. The tool supports strict exact-order funnels, unordered funnels, and allows teams to specify the exact time window users have to convert. A massive competitive advantage is its seamless integration with the Session Replay feature; analysts can instantly click on the segment of users who dropped off at step 3 and watch the actual video recordings of those specific users struggling. This provides an immediate qualitative explanation for the quantitative drop-off, a capability rarely found in traditional analytics suites.
The Paths tool visualizes organic user navigation, allowing teams to map journeys based on pageviews, custom events, or screen names.
To understand how users naturally navigate an application, the platform provides a dynamic User Paths visualization. This tree-graph maps out the exact sequence of events or pageviews users trigger before or after a specific milestone. A key strength is the ability to heavily customize the nodes; analysts can choose to only map specific custom events, ignore noisy pageviews, or group URLs by regex rules to clean up the visual flow. This makes it highly effective for analyzing complex Single Page Applications (SPAs). While incredibly useful for spotting navigation loops, highly complex enterprise apps with thousands of events can quickly render the visualization overwhelming without strict filtering.
The platform provides robust cohort and retention analysis, allowing teams to group users by complex behavioral triggers over customized timeframes.
Retention tracking is built deeply into the platform, allowing product teams to analyze long-term user loyalty. Analysts can define specific behavioral cohorts (e.g., "users who completed onboarding and used the core feature twice") and track how they return over subsequent days, weeks, or months. The interface supports both standard N-day retention and bracketed retention, which is crucial for products that don't expect daily usage (like a monthly tax software). Crucially, these defined cohorts can be saved dynamically and used not just for reporting, but directly as targeting segments for the platform's native Feature Flags and A/B testing modules.
Anomaly detection automatically highlights statistical outliers in event trends and correlates them directly with qualitative session recordings.
The platform features native anomaly detection that automatically analyzes historical event volumes to establish expected trend baselines. When a metric deviates significantly—such as a sudden collapse in successful checkouts or a massive spike in API errors—the system visually highlights the anomaly on the chart. What sets this apart from competitors is the immediate next step: when an anomaly is flagged, analysts can click the outlier point to instantly watch the session recordings of the users affected during that specific timeframe. This drastically reduces the time it takes to diagnose whether a data spike is a tracking error, a backend bug, or genuine viral traffic.
While it lacks predefined retail templates, e-commerce workflows can be deeply analyzed using its flexible custom event schema and conversion funnels.
Because the platform is engineered as a general-purpose product analytics suite, it does not provide native, plug-and-play e-commerce dashboards (like built-in "Revenue" or "Cart Abandonment" templates). Instead, online retailers must instrument the purchase funnel manually using custom events (e.g., Add to Cart, Purchase) and pass transaction amounts as event properties. Once configured, the platform excels at analyzing the deeply complex behavioral paths that lead to a sale, combining revenue data with session recordings to see exactly why a user abandoned a cart. However, teams looking for instant, out-of-the-box retail metrics will find the initial setup more demanding than dedicated e-commerce analytics tools.
The platform handles cross-device tracking by allowing developers to explicitly map anonymous session IDs to authenticated internal user IDs.
Identity resolution is managed through a deliberate, developer-controlled merging process using the identify and alias API calls. When a user browses anonymously, they are assigned a distinct ID. Once they authenticate (log in), developers trigger an identify call with the user's permanent backend database ID. The system then automatically stitches the pre-login anonymous behavior with the authenticated profile, maintaining a continuous historical journey across different devices and browsers. This ensures highly accurate tracking for SaaS and subscription businesses. However, this deterministic approach relies entirely on users actually logging in; it does not attempt to guess identity using opaque third-party data graphs.
Custom data modeling operates entirely on an event-driven schema, allowing for unlimited custom event definitions rather than relying on rigid session tracking.
The platform discards the rigid, session-centric model of legacy web analytics in favor of a profoundly flexible event-driven architecture. Every action a user takes is logged as an independent event associated with their profile. This model allows data architects to define virtually any custom interaction and attach extensive JSON metadata properties to it. Furthermore, the platform supports "Group Analytics," which allows B2B companies to track behavior at an account or company level rather than just at an individual user level. This flexibility is incredibly powerful for complex SaaS products but requires organizations to maintain strict data governance to prevent the taxonomy from becoming chaotic.
The Data Management feature provides a centralized dictionary to verify, annotate, and govern custom events and properties across the organization.
To help teams maintain a clean and reliable event schema, the platform includes a native Data Management module. This acts as a centralized data dictionary where administrators can add descriptions to specific events, verify that they are firing correctly, and tag them to specific product owners. If the tracking taxonomy becomes bloated, analysts can use this interface to hide obsolete events or merge duplicate tracking codes without having to deploy new code. While it provides solid foundational governance, it lacks the highly aggressive, automated event-quarantine features found in top-tier enterprise platforms that automatically block malformed data payloads before they hit the database.
The experimentation framework uniquely integrates A/B testing and multivariate experimentation directly into the analytics platform alongside feature flags and session replay.
A massive differentiator for this platform is its native inclusion of a full-fledged A/B testing and experimentation engine. Teams can launch A/B tests or multivariate experiments directly from the UI, utilizing the platform's native Feature Flags to split traffic. Because the testing engine shares the exact same database as the analytics engine, analysts can evaluate experiment results using complex, long-term behavioral metrics (like 30-day retention or downstream feature usage) rather than just basic click-through rates. This eliminates the severe data discrepancies that typically occur when trying to sync a standalone third-party testing tool with a separate analytics platform.
The platform offers EU data residency and open-source deployment options to ensure strict compliance with global privacy regulations.
The platform provides robust options for ensuring GDPR and CCPA compliance. For cloud customers, it offers dedicated EU data residency, ensuring that European user data never leaves European servers. However, its strongest compliance feature is its open-source nature; organizations with ultra-strict privacy requirements (like healthcare or finance) can self-host the platform entirely on their own infrastructure, ensuring no third-party vendor ever touches their behavioral data. It also includes native tools for anonymizing IP addresses and processing specific user data deletion requests. Like all analytics tools, legal compliance still depends heavily on the business correctly implementing a Consent Management Platform (CMP) on their frontend.
The Data Pipelines feature allows seamless, automated export of raw, unsampled event data to external warehouses like BigQuery or S3.
For organizations needing to centralize their data architecture, the platform features native Data Pipelines (formerly known as apps or plugins). These pipelines allow teams to configure automated, continuous streams of raw, hit-level JSON data directly into major data warehouses such as Google BigQuery, Snowflake, Amazon S3, or Redshift. This enables data science teams to easily combine product behavioral data with external financial or CRM datasets. Because the platform does not artificially restrict raw data access or mandate aggressive data sampling, this feature provides total data portability and ownership, a critical requirement for enterprise data engineering teams.
Data retention policies are flexible, with standard cloud plans offering 1 to 7 years of retention, and self-hosted instances providing limitless storage.
Data retention limits on this platform depend heavily on the chosen deployment method. For the managed Cloud version, standard retention for granular event data ranges from 1 year to 7 years depending on the specific premium plan, ensuring deep historical analysis for long-term product trends. However, organizations that choose to self-host the open-source version on their own infrastructure have absolute control over their data retention; they can retain unsampled, user-level data indefinitely, limited only by their own server capacity. The platform also includes administrative tools to automatically purge old data if an organization needs to enforce strict data minimization rules for legal compliance.
This mobile analytics capability provides robust support for mobile app measurement via dedicated SDKs, unifying mobile behavior, feature flags, and session replays.
The platform excels at mobile app analytics by natively unifying several different capabilities into its mobile SDKs (iOS, Android, React Native, Flutter). Instead of just tracking events, these SDKs allow mobile teams to simultaneously deploy feature flags to control mobile rollouts and record in-app user sessions (session replay for mobile is supported on select frameworks). This provides product teams with a holistic view of the mobile experience that standard analytics tools cannot match. Because the platform uses a flexible event-driven schema, it perfectly accommodates the complex, state-based nature of mobile applications, making it a strong alternative to legacy mobile trackers.
Secure access is managed through native SAML 2.0 Single Sign-On, allowing seamless integration with corporate identity providers like Okta or Azure AD.
To support enterprise deployments, the platform natively features SAML 2.0 Single Sign-On (SSO) integration. This allows IT administrators to connect the analytics environment directly to centralized corporate identity providers such as Okta, Microsoft Entra ID (Azure AD), or Google Workspace. Implementing SSO enables organizations to enforce multi-factor authentication, automate user provisioning, and quickly revoke access when team members depart. This centralized access control is critical for maintaining data security and compliance within large organizations, eliminating the vulnerabilities associated with managing standalone passwords within the analytics application itself.
Users can build highly complex, dynamic behavioral cohorts and instantly use them to target A/B tests or feature flags.
The segmentation engine allows analysts to create highly granular "Cohorts" based on specific user properties, historical event frequencies, and strict time constraints. A major competitive advantage is how these segments are activated. Because the platform includes native Feature Flags and Experimentation, these complex behavioral cohorts can be used instantly to target specific app experiences. For example, a product manager can build a cohort of "users who experienced 3 errors yesterday" and immediately deploy a feature flag that offers them a specific apology discount in-app. This creates a seamlessly integrated workflow from deep behavioral analysis directly to personalized product activation.