OpenPanel

Open-source product analytics, honestly reviewed. No marketing fluff, just what you get when you self-host it.

TL;DR

• What it is: Open-source (AGPL-3.0) web and product analytics platform — think Mixpanel, but the source code lives on your server [README].
• Who it’s for: Non-technical founders and small product teams who want event-level product analytics (funnels, cohorts, user profiles) without Mixpanel’s per-event billing or GA4’s opaque data model [homepage].
• Cost savings: Mixpanel’s Growth plan starts around $28/mo and scales sharply with event volume. OpenPanel self-hosted runs on a VPS for the cost of the hardware — exact cloud pricing tiers weren’t publicly accessible at time of writing, but the site advertises a 30-day free trial with no credit card required [homepage].
• Key strength: Unusually complete feature set for an open-source tool — funnels, cohorts, user profiles, session replay, A/B testing breakdowns, and real-time dashboards are all included, not gated behind separate paid add-ons [README].
• Key weakness: AGPL-3.0 license means you can self-host freely, but embedding OpenPanel in commercial software or distributing a modified version requires your product to also be AGPL. Not MIT. Read the license before you embed it [README]. Setup is also heavier than single-binary tools — it requires Postgres, ClickHouse, and Redis running in concert.

---

What is OpenPanel

OpenPanel is a product analytics platform. You instrument your app with its SDK, and in return you get a dashboard showing who visited, what they clicked, where they dropped off in a signup flow, whether the users who completed onboarding retained better than those who skipped it, and what a specific user’s full session looked like — including a video replay of exactly what they did on screen.

The project describes itself as “an open-source alternative to Mixpanel” that “combines the power of Mixpanel with the ease of Plausible and one of the best Google Analytics replacements” [README]. That’s a broader pitch than most self-hosted analytics tools make, and it’s mostly accurate: most open-source analytics tools pick a lane — either simple page-view counting (Umami, Plausible) or complex event pipelines (PostHog). OpenPanel is trying to hold both ends: Plausible-level ease of setup, Mixpanel-level analytical depth.

The project sits at 5,539 GitHub stars under AGPL-3.0 licensing [merged profile]. It was built by Carl Lindesvard and has been running in production long enough to be trusted by over 1,000 projects according to the homepage [homepage]. The technical stack is explicit in the README: Next.js for the dashboard, Fastify for the event API, Postgres for metadata storage, ClickHouse for event storage, Redis for caching and queues, BullMQ for job processing, tRPC for the API layer, and Tailwind + shadcn/ui for the interface [README]. That’s a real, modern production stack — not a toy project.

---

Why people choose it

Most open-source analytics tools force a trade-off. Plausible is clean and private, but you get aggregate page views and simple goals, not per-user event timelines or funnel analysis. PostHog is feature-complete, but it’s complex to deploy, the free tier has limits, and its open-source version has quietly drifted away from the hosted product. Mixpanel gives you everything, but you’re paying per-event on someone else’s servers.

OpenPanel is positioned at the gap: product-analytics depth, self-hosting, GDPR compliance by default, and a UI that real users describe as easier than the incumbents.

Homepage testimonials give a ground-level view. Julien Hany of Strackr.com: “After testing several product analytics tools for Strackr, we chose OpenPanel and we are very satisfied with the product. We have been using it since the beta, and there are constant updates. Profiles and Conversion Events are our favorite features.” [homepage] Thomas Sanli of suneed.best: “Before OpenPanel, I was using Plausible, and it was ok. But OpenPanel is like 10 leagues ahead!! Better UX/UI, many more features, and incredible support from the founder. Bonus point: it’s 1 click install on Coolify!!” [homepage]

Twitter/X gives a similar picture. Steven Tey, a developer with a large following, wrote: “Open-source Mixpanel alternative just dropped → It combines the power of Mixpanel + the ease of use of Plausible into a fully open-source product” [homepage]. Piotr Kulpinski noted: “The Overview tab in OpenPanel is great. It has everything I need from my analytics: the stats, the graph, traffic sources, locations, devices, etc. The UI is beautiful ✨ Clean, modern look, very pleasing to the eye.” [homepage]

The pattern is consistent: people come from Plausible because they need more depth, or from Mixpanel/GA4 because they want out of per-event billing and data-sovereignty concerns. Nobody in the available testimonials came from PostHog, which suggests these are two different audiences — OpenPanel is pulling from the “I need Mixpanel features but I’m not an enterprise” segment, not from engineers who wanted PostHog but cheaper.

---

Features

What the README and homepage describe as available — self-hosted, not gated behind cloud tiers:

Web and product analytics:

• Pageview and visitor tracking, referrer attribution, device and location breakdowns [README]
• Custom event tracking with properties [README]
• Real-time dashboards with live visitor counts [README][homepage]
• Flexible chart types: line, bar, Sankey flows, custom combinations [homepage]

User-level analytics:

• Individual user profiles with full event history and session timeline [README]
• Cohort analysis: group users by behavior and see how they evolve over time [README]
• Funnel analysis: define a sequence of events, see where users drop off [README]
• Retention analysis: day 0, day 1, day 7, day 30 cohort retention table [homepage]

Session intelligence:

• Session replay: full video recording of user sessions with privacy controls built in [README][homepage]
• Rage click detection, scroll tracking [homepage]
• Notably: the feature comparison table in the README shows Mixpanel caps session replay at 5,000 sessions/month on free and 20,000 on paid — OpenPanel claims no such limit [README]

Product and growth tooling:

• A/B testing: instrument variants and break down conversion by experiment arm [README][homepage]
• Revenue tracking: connect payment events, track MRR, see which referrer channels drive actual revenue [homepage]
• Google Search Console integration: see which search queries convert [homepage]
• Smart notifications: get alerted when a funnel completes, not just when traffic spikes [README][homepage]

Developer features:

• SDKs for Web, iOS (Swift), Android (Kotlin), React Native [README]
• Server-side tracking [README]
• Webhooks and event forwarding to third-party systems [homepage]
• Full API access [README]

Privacy and compliance:

• Cookieless tracking by default [README][homepage]
• GDPR-compliant, EU-hosted (for the managed cloud version) [homepage]
• No data sent to third parties when self-hosted

The feature comparison table in the README maps OpenPanel against Mixpanel, GA4, and Plausible across 15 dimensions and claims full marks across the board [README]. Take that with appropriate skepticism — it’s self-reported — but the individual feature descriptions in the README and homepage are specific enough to be credible, not vague marketing claims.

---

Pricing: SaaS vs self-hosted math

OpenPanel Cloud: The homepage advertises a 30-day free trial with no credit card required and describes “simple, transparent pricing,” but the actual tier structure and prices weren’t publicly visible in the scraped content at time of writing. The site claims “no hidden costs or usage limits” [README]. For exact current pricing, check https://openpanel.dev directly.

Self-hosted (AGPL-3.0): The software is free. Your costs are:

• A VPS: $5–20/month depending on event volume. ClickHouse is the heavy component — it’s fast, but it needs memory. Budget at least 4GB RAM for anything beyond hobby use.
• Time to set up and maintain

Mixpanel for comparison:

• Free: 20M events/month, limited reports
• Growth: starts around $28/mo, scales with event volume — heavy apps can hit hundreds per month quickly
• Enterprise: custom pricing

GA4: Free, but raw data access requires BigQuery, which charges by query/storage. GA4 has also faced GDPR enforcement actions in multiple EU countries [README feature comparison table].

Plausible: $9/mo for up to 10,000 pageviews, scaling to $19/mo (100k), $29/mo (200k), etc. Simpler product, simpler pricing — but you’re giving up funnels, cohorts, user profiles, and session replay.

The self-hosting math is straightforward: if you’re paying Mixpanel $50–200/mo and you have someone technical enough to run Docker Compose, the $6–12/mo VPS calculates favorably. The real cost is deployment time and ongoing maintenance. If you’re a solo non-technical founder and you’ve never SSH’d into a server, that’s the actual barrier — not the feature set.

---

Deployment reality check

OpenPanel is not a single-binary tool. The production setup requires Postgres, ClickHouse, Redis, and the application itself — four components that all need to run and stay healthy [README]. The README ships a docker-compose setup, and the development instructions are clean:

pnpm install
cp .env.example .env
pnpm dock:up
pnpm migrate:deploy
pnpm dev

That’s reasonable for a developer. For a non-technical founder, this is a meaningful lift.

What you actually need:

• A Linux VPS with 4GB+ RAM (ClickHouse is memory-hungry; 2GB will struggle)
• Docker and docker-compose
• A domain with HTTPS (nginx or Caddy reverse proxy)
• An SMTP provider for user invite emails
• Basic comfort with reading server logs when something breaks

What can go wrong:

• ClickHouse is production-grade and powerful, but it adds operational complexity that simpler analytics tools (Umami, Plausible) don’t have. If it crashes, you need to know how to restart it.
• The AGPL license is a deployment-stopper for some companies. If your legal team hasn’t reviewed what AGPL means for your SaaS stack, do that before deploying.
• Homepage testimonial mentions “1 click install on Coolify” [homepage] — if you’re running Coolify already, that path is significantly easier than a raw VPS setup.

Realistic time estimate: 1–2 hours for a developer on a fresh VPS. For a non-technical founder with Coolify already running: 15–30 minutes per the testimonial. Without Coolify, budget a full afternoon plus troubleshooting time.

---

Pros and Cons

Pros

• Feature-complete out of the box. Funnels, cohorts, user profiles, session replay, A/B testing, real-time dashboards, revenue tracking — all included, not split across tiers [README]. Comparable to what Mixpanel charges significantly for.
• Cookieless and GDPR-compliant by default. No consent banner required, no legal grey zone with EU regulators [README][homepage]. GA4 has been banned in multiple EU countries over cross-border data transfers; OpenPanel self-hosted has no such risk.
• Session replay with no monthly cap. Mixpanel limits replay to 5K/month free and 20K/month paid. OpenPanel claims no limit [README]. For a product team doing qualitative UX research, that matters.
• Clean, modern UI. Multiple independent testimonials call out the UI specifically — not as a throwaway compliment but as a primary reason they stayed [homepage testimonials].
• Active development. Multiple users note “constant updates” and strong founder support [homepage]. The project has real momentum at 5,539 stars.
• Transparent tech stack. Postgres + ClickHouse + Redis is a well-understood combination. You can hire engineers who know these, migrate data out, or extend the system without vendor-specific knowledge [README].
• Revenue and SEO data in one place. The ability to connect payment events and Search Console to the same dashboard as behavioral analytics is uncommon at this price point [homepage].

Cons

• AGPL-3.0 is not MIT. The license allows free self-hosting, but if you distribute modified versions or embed OpenPanel in a commercial product, your own code may need to be AGPL as well. For a SaaS founder building on top of OpenPanel, read the license carefully [merged profile].
• Heavy deployment stack. Four components (app + Postgres + ClickHouse + Redis) require more operational overhead than single-binary tools. ClickHouse in particular has a steeper learning curve than PostgreSQL if things go wrong [README].
• Exact pricing unavailable. The cloud pricing tiers weren’t accessible at time of review. “Transparent pricing” is claimed on the homepage but the actual numbers require visiting the site [homepage].
• Smaller ecosystem than PostHog or Mixpanel. No plugin marketplace, no third-party integrations built on the platform — just webhooks and event forwarding. If you need tight Salesforce or HubSpot integration, you’re building it yourself [homepage].
• Single creator origin. The project appears to be primarily one founder’s work (Carl Lindesvard). That’s not inherently a problem — many excellent open-source projects start this way — but it’s a succession risk for a mission-critical analytics stack. No data on team size or contributor count in the available sources.
• No documented REST API surface for programmatic flow management. Beyond SDK instrumentation and webhook forwarding, there’s no evidence of a mature API for external systems to query your analytics data [available sources]. If you need to build dashboards in your own product that pull from OpenPanel, that path is unclear.

---

Who should use this / who shouldn’t

Use OpenPanel if:

• You’re currently paying Mixpanel $50–200/mo and the only thing stopping you from self-hosting is finding a tool with comparable depth.
• You need funnels, cohorts, user profiles, and session replay — and you’re not willing to pay Mixpanel’s rates for them.
• You want GDPR compliance without a consent banner and without relying on a US-based analytics provider.
• You’re running Coolify or another Docker-management platform and want a one-click-ish deployment path.
• You can accept AGPL licensing (no plans to embed or redistribute a modified version commercially).

Skip it (pick Plausible instead) if:

• You only need page views, traffic sources, and basic goals — not per-user event tracking.
• You want the simplest possible setup with the lowest operational footprint.
• You’re non-technical and don’t have anyone to manage a multi-container Docker setup.

Skip it (pick PostHog instead) if:

• You need a mature plugin ecosystem and extensive third-party integrations.
• You have a large engineering team already comfortable with PostHog’s feature flag and experimentation system.
• You want a MIT-licensed codebase for embedding purposes.

Skip it (stay on Mixpanel) if:

• Your compliance team requires SOC 2 Type II with a vendor you can put in your vendor risk list.
• You have no technical resources to manage infrastructure and need fully managed analytics with SLAs.
• You’re running at massive event volume and need Mixpanel’s enterprise support guarantees.

Skip it (pick Umami) if:

• You need a truly lightweight, single-binary, low-maintenance analytics tool and don’t need product analytics depth.

---

Alternatives worth considering

• Plausible — simpler, single-binary, MIT-licensed, $9/mo cloud. Gives you page views and basic goals. Lacks funnels, cohorts, user profiles, session replay. The tool OpenPanel users frequently say they outgrew.
• PostHog — the closest comparable in the open-source product analytics space. More integrations, larger ecosystem, feature flags, A/B testing at scale. MIT-licensed core. Heavier deployment, more complex UI, faster-moving product surface. Worth evaluating head-to-head with OpenPanel.
• Umami — lightweight, privacy-first, MIT-licensed page-view analytics. No event-level product analytics. Best for content sites that don’t need funnel tracking.
• Metabase + your own event table — if you already have engineers and a data warehouse, you can build equivalent dashboards yourself. Requires more work, but gives you full control over the schema.
• Mixpanel — the incumbent. Fully managed, largest feature set, most expensive at scale, closed source. The tool OpenPanel explicitly benchmarks against.
• GA4 — free but legally problematic in the EU, requires BigQuery for raw data, has a notoriously poor UX for product analytics workflows.
• Fathom — privacy-first, GDPR-compliant, cleanly designed cloud-only tool. No self-hosting option. Simple page-view analytics, not product analytics.

---

Bottom line

OpenPanel is the most feature-complete open-source product analytics tool in its niche. If you want funnels, cohorts, session replay, and real-time dashboards without Mixpanel’s per-event billing or GA4’s legal exposure in the EU, it’s the strongest self-hosted candidate available right now. The AGPL-3.0 license and four-component deployment stack are real constraints — not showstoppers for most founders, but things you need to know before committing. The UI and UX get consistently praised by users who moved from both Plausible (too simple) and Mixpanel (too expensive), which puts it squarely in the gap where most non-technical founders actually live.

If the deployment is the blocker, that’s exactly what upready.dev sets up for clients. One-time fee, done, you own the infrastructure and the data.

---

Sources

1. OpenPanel GitHub README — feature list, stack, deployment instructions, license (AGPL-3.0), 5,539 stars. https://github.com/openpanel-dev/openpanel
2. OpenPanel homepage — testimonials, feature overview, comparison table, pricing claims. https://openpanel.dev
3. OpenPanel merged profile — structured metadata: slug, category, stars, license, canonical features. Internal pipeline data.

Note: Third-party review articles were unavailable or inaccessible for this tool at time of writing. This review is based on primary sources (GitHub README, official website, homepage testimonials). Independent third-party assessments should be sought before making infrastructure decisions.