Open-source Next.js analytics dashboards compared (2026)

Data Lighthouse

This is a self-hosted Next.js template, not a hosted analytics product. The dashboard uses shadcn/ui and Recharts. Sessions and customers are Postgres tables defined with Drizzle. A Supabase Postgres URL works, including the transaction pooler. The repository does not include supabase-js, and sign-in is Clerk with GitHub OAuth. The template README documents a Vercel deploy. There is no Dockerfile and no ClickHouse dependency here. The tracker stores a visitor id in localStorage, so it is not the cookieless default Insight and OpenAnalytics describe.

Stripe is optional and reads STRIPE_SECRET_KEY on your deployment. The overview cards are revenue for 7, 14, or 30 days, average daily revenue, MRR from active subscriptions normalized to one month, ARR as that MRR times 12, and revenue per visitor from distinct Postgres visitors in the same window. Definitions are on the Next.js Stripe analytics dashboard page. shadcn and the Drizzle schema are on the shadcn Next.js Supabase analytics dashboard page. The license is $59 once ($99 compare-at on the pricing page), paid on Polar after GitHub sign-in. Stripe is not the checkout processor for the license.

rrHog

Choose rrHog when session replay is the point. The README describes an MIT, self-hosted pipeline: a Next.js UI, FastAPI, NATS JetStream, ClickHouse for events, Postgres for metadata, and rrweb. The quick start is Docker Compose with nginx, NATS, ClickHouse, Postgres, the API, a worker, and the web app. It does not mention Stripe or revenue. That is a better fit than Data Lighthouse if you will run those services and you need replay. It is a separate system, not a dashboard you paste into an existing Next.js app.

Insight

Choose Insight when you want web analytics and Stripe revenue with no license fee. The README says MIT, Next.js 16, React 19, Tailwind CSS v4, ClickHouse, SQLite, and Docker Compose on a Linux server with 1 vCPU and 2 GB of RAM. The default tracker stores nothing on the visitor's device. Stripe is a read-only key: revenue net of refunds, a conversion rate, and revenue per visitor. The same README lists funnels, retention, and AI crawler fetches. It does not list MRR or ARR. Insight is the stronger free choice if a small VPS is acceptable. It is a server you operate, not shadcn source inside your product, and it does not run as only Vercel plus Supabase.

OpenAnalytics

Choose OpenAnalytics when you want a fuller privacy-first product and you can host it. The README is AGPL-3.0 for the server and MIT for the browser tracker. The dashboard is Next.js. Stores are Postgres, ClickHouse, and two Valkey instances. Self-hosting asks for Docker, about 4 GB of RAM, 25 GB of disk, and four DNS names. Features named in the README include page views, custom events, sessions, web vitals, funnels, retention, revenue from your Stripe account, and an MCP server. A hosted instance runs at getopen.so. That page's price was not in the README, so it is not quoted here. The AGPL's network-use terms apply if you run a modified copy as a service. That is a better fit than a $59 template when you need those features and will run the cluster.

Tinybird Web Analytics starter kit

Choose this kit when Tinybird is where events should live. The repository is MIT. You deploy a Tinybird data project, send page_hit events with flock.js, and can track custom events and web vitals. The dashboard is Next.js. Its README says React 18, Tailwind, and SWR, and documents a Vercel deploy. package.json pins next 14.2.35. Local development uses Tinybird Local. The README does not mention Stripe, and it does not ask you to operate ClickHouse yourself. A Tinybird usage price was not in the README. Use it instead of Data Lighthouse when a Tinybird workspace is the backend you want. Use Data Lighthouse when the database should be your own Postgres.

Open SaaS Board

Choose Open SaaS Board when the job is a free board on top of a Supabase project and Stripe or Lemon Squeezy. The README says Next.js 16, React 19, Tailwind, shadcn/ui, and Supabase Auth. Setup is npm, a SQL file in the Supabase editor, and an onboarding wizard the README calls five minutes. API keys stay in localStorage. It lists revenue, active subscriptions, new subscribers, churn, and user analytics. It asks "what's my MRR?" and does not publish the formula. It does not describe a pageview tracker. The README says MIT, but on 2026-11-03the repository root had no LICENSE file and GitHub's license field was empty, so treat that claim as unverified. It also recommends a Netlify URL while the roadmap still lists a cloud version as coming soon. It is the better free choice when you already have those accounts and you do not need first-party web analytics.

SaaS Monitor

Choose SaaS Monitor when you want Stripe on a screen: MRR or ARR, revenue today and this month, customers, net new MRR, sounds, and celebrations. The README says MIT, Next.js 16, React 19, Tailwind CSS 4, shadcn/ui, Drizzle, MySQL, and Better Auth. Self-hosting is Docker Compose, including a Raspberry Pi. The hosted app is described as free during the public beta. It does not describe website visitor analytics. It is the better free choice for an office TV of Stripe numbers. It is not a substitute for a traffic dashboard.

Next.js SaaS Starter

Choose the Next.js SaaS Starter at github.com/nextjs/saas-starter when you are building the SaaS itself. The README lists a marketing page, Stripe Checkout, user and team CRUD, owner and member roles, the Stripe Customer Portal, email and password auth, and activity logging. The stack in the README is Next.js, Postgres, Drizzle, Stripe, and shadcn/ui. Production steps are a Vercel deploy and POSTGRES_URL. The README does not mention ClickHouse, Docker Compose, pageview analytics, or an MRR card. GitHub reports the license as MIT. package.json listed next 15.6.0-canary.59. This is the better free start for auth and billing. It is not the analytics dashboard those searches are asking for.