InsightFlare

InsightFlare records visits with a tracking script of about 3 KB gzipped and does not store raw IP addresses. Its README describes the tool as GDPR compliant and says visits can be tracked without asking for consent.

Geolocation comes from Cloudflare, and a tracking intensity mechanism automatically adjusts how long visitor identifiers persist based on regional privacy rules. The SDK also records performance metrics and captures custom events through data-insightflare-event attributes or a JavaScript API. Its identify() call links visits to a signed-in user, and reset() ends that identity at logout. The multilingual dashboard reports real-time visitors, pages and traffic sources, UTM campaigns, individual sessions and visitors, funnels and JSON event properties. Teams can also publish public dashboards, issue scoped API keys and connect AI agents through a skills file at /.well-known/skills.json. Scheduled and conditional email notifications go out through Resend, with the API key entered in the system settings. A single Worker, built with TanStack Start and the Cloudflare Vite plugin, ingests events through a Durable Object, stores them in D1, keeps site settings in KV and runs a cron trigger every 30 minutes. The Deploy to Cloudflare button clones the repository, provisions D1 and KV, and asks for MAIN_SECRET and BOOTSTRAP_ADMIN_PASSWORD. About three minutes later you can sign in as admin at the workers.dev URL. An optional GitHub App opens a pull request in your deployment repository when upstream releases a new version.

Built for: Website owners, indie founders and small product teams who want self-hosted, cookie-free analytics on Cloudflare.

What you can use it for

• Measure site traffic without cookies or stored IP addresses
• Track signup and pricing-page events with funnels and UTM reports
• Share a public traffic dashboard with clients or users
• Give an AI agent access to analytics through scoped API keys
• Send scheduled traffic reports by email

Deploy

• One-click · main path

  Branch main

  Deploy to Cloudflare

• Manual

  Branch main

  Fork the repository, create a D1 database and a KV namespace, add their IDs to wrangler.toml and set the variables from .dev.vars.example, then import the fork from the Workers dashboard.

Setup is the deploy button plus two secrets, MAIN_SECRET and BOOTSTRAP_ADMIN_PASSWORD.

What it needs to run

• Analytics Engineinsightflare_event_facts, insightflare_request_events, insightflare_traffic_events
• D1 databasesinsightflare
• Durable ObjectsIngestDurableObject
• KV namespacesSITE_SETTINGS_KV
• R2 bucketsinsightflare-archive

Good to know

Data older than one year is compressed and cannot be filtered unless you create an R2 bucket and enable the commented-out archive binding, which the deploy button does not set up. Analytics Engine must be enabled by hand in the dashboard, and reading its datasets needs an API token with Account Analytics read permission.

Cost on Cloudflare

Where this comes from

• Repositoryhttps://github.com/RavelloH/InsightFlare
• App websitehttps://demo.insightflare.net

An alternative to Amplitude, Clicky, Fathom Analytics, GoatCounter, Google Analytics, Matomo, Mixpanel, Plausible Analytics, PostHog Web Analytics, Simple Analytics, Statcounter, Umami, Vercel Web Analytics.