Attribution & Analytics
PostHog
PostHog is a developer-focused product platform that combines analytics, experimentation, observability, and data tools. Its products include product and web analytics, session replay, heatmaps, feature flags, experiments, surveys, error tracking, logs, data warehouse connections, CDP workflows, and AI-assisted investigation, with cloud hosting and self-managed deployment options for supported components. It suits product engineers and technical teams that want behavior analysis and release controls in one stack, especially when transparent pricing, APIs, and data ownership matter.
What it does
PostHog gives product and growth teams an integrated stack for understanding what users do after they convert, which makes it useful for attribution work that extends past the click. Event autocapture and SDKs record behavior, then funnels, retention, paths, and cohort tools tie acquisition sources to downstream activation and revenue. Session replay and heatmaps add qualitative context, while feature flags and experiments let teams test changes and measure causal lift rather than correlation. A built-in data warehouse and CDP-style transformations let you join ad-platform and CRM data against product events. Available as cloud or self-hosted, with transparent usage pricing, it appeals to engineering-led teams that want ownership of their behavioral data and full-funnel measurement in one place.
Where it fits
PostHog works at the analytics end of attribution, connecting acquisition events to activation and retention so teams see which sources drive durable value, not just signups.
Core features
- Event autocapture with funnels, retention, and path analysis
- Session replay and heatmaps for qualitative context
- Feature flags and A/B experiments for causal testing
- Data warehouse and CDP transformations for source joins
- Cloud or self-hosted deployment with usage-based pricing
Best for
- Engineering-led teams wanting full-funnel behavioral measurement
- Products needing experimentation and analytics in one stack
- Teams that require data ownership or self-hosting
Beginner notes
- Define a small set of core events and a north-star conversion before instrumenting everything, or autocapture noise buries the signals that matter.
- Set UTM and source properties on signup events so you can build attribution funnels later without reprocessing history.
- Watch your event volume on usage pricing; sampling high-frequency client events keeps costs predictable as traffic grows.