🩺 Vitals
- 📦 Version: posthog-cli-v0.7.2 (Released 2026-03-17)
- 🚀 Velocity: Active (Last commit 2026-03-19)
- 🌟 Community: 32.1k Stars · 2.4k Forks
- 🐞 Backlog: 3238 Open Issues
🏗️ Profile
- Official: posthog.com
- Source: github.com/PostHog/posthog
- License: MIT (Core) / ELv2 (Enterprise)
- Deployment: Cloud /Docker
- Data Model: ClickHouse (OLAP) + PostgreSQL
- Jurisdiction: USA 🇺🇸
- Compliance: SOC 2 Type II, HIPAA, GDPR, CCPA
- Complexity: High (5/5) - Distributed ClickHouse Architecture
- Maintenance: High (4/5) - Requires Data Engineering
- Enterprise Ready: High (5/5) - Scalable to Billions of Events
1. The Executive Summary
What is it? PostHog is an integrated "Product OS" that consolidates multiple fragmented tools—product analytics, session replay, feature flagging, and A/B testing—into a single platform. For CTOs, it solves the "data silo" problem where user behavior data (Analytics) is disconnected from the control plane (Feature Flags). By combining these, you can instantly see how a new feature rollout impacts user retention or revenue without complex data pipelining.
The Strategic Verdict:
- 🔴 For Simple Websites: Overkill. If you just need pageviews, use Plausible or Umami. PostHog is for applications where user behavior is complex.
- 🟢 For SaaS Products & Scale-Ups: Strong Buy. It replaces a stack that would cost $50k+/year (Amplitude + LaunchDarkly + Hotjar) with a unified, cheaper, and more cohesive open-source alternative.
2. The "Hidden" Costs (TCO Analysis)
| Cost Component | Amplitude (SaaS) | PostHog (Self-Hosted) |
|---|---|---|
| Tool Stack | 3+ contracts (Amplitude, LD, Hotjar) | 1 unified platform |
| Data Ingress | Paying to send data to 3 places | Send once, use everywhere |
| Infrastructure | N/A (SaaS only) | High (ClickHouse is complex) |
3. The "Day 2" Reality Check
🚀 Deployment & Operations
- Installation: PostHog creates a distinction between "Hobby" deployment (Docker Compose, good for testing) and "Scale" deployment (requires Kubernetes and careful ClickHouse management).
- Scalability: Extremely high. Built on ClickHouse, it can handle billions of events. However, self-hosting at scale requires dedicated Data Engineering resources.
🛡️ Security & Governance
- Access Control: Granular project-level and organization-level permissions.
- Data Handling: PostHog enables full data sovereignty. You can host it in your own VPC (AWS/GCP) so user data never leaves your infrastructure, a critical requirement for Fintech and Healthcare.
4. Market Landscape
🏢 Proprietary Incumbents
- Amplitude
- LaunchDarkly
- Hotjar
- Mixpanel
🤝 Open Source Ecosystem
- Matomo: The industry standard for privacy-focused web analytics; use Matomo for SEO and marketing, and PostHog for product-led growth and behavioral experimentation.
- Umami: A lighter alternative if you only need simple, cookieless web analytics.