Ackee

Self-hosted website analytics, honestly reviewed. No marketing fluff, just what you get when you run it on your own server.

TL;DR

• What it is: Self-hosted, Node.js + MongoDB analytics tool that tracks website traffic without cookies, without unique user identification, and without sending data anywhere except your own server [README].
• Who it’s for: Developers, indie founders, and privacy-conscious site owners who want traffic insights — page views, referrers, durations — without running Google Analytics or Matomo. Not for growth teams who need funnels, cohorts, or A/B testing.
• Cost savings: Google Analytics is free but costs you your visitors’ data. Matomo Cloud starts at $19/mo. Plausible starts at $9/mo. Ackee self-hosted runs on a $5–10/mo VPS with unlimited domains and zero per-site fees [README][1].
• Key strength: Genuinely cookie-free by design, GDPR-compliant without a consent banner, minimal interface that doesn’t try to be everything, and a GraphQL API that lets you build on top of it [2][README].
• Key weakness: This is a solo-developer project with limited active maintenance. No funnels, no session recordings, no user journeys, no A/B testing — and if you need any of that, Ackee isn’t the answer. The ecosystem of integrations is small and some (like the WordPress plugin Soapberry [3]) haven’t been updated in years.

---

What is Ackee

Ackee is a self-hosted analytics tool built on Node.js and MongoDB. You deploy it on your own server, add a tracking script to your site (identical workflow to the Google Analytics snippet), and it starts collecting traffic data: page views, unique visitor counts, session durations, referrers, language, screen size, OS, browser, and device type [README].

The name comes from the ackee fruit, and the project tagline — “Track without being tracked” — is actually accurate rather than marketing copy. Ackee doesn’t use cookies. It doesn’t track individual users across sessions. It applies a multi-step anonymization process so you can see aggregate traffic patterns without storing anything that makes a visitor individually identifiable [2][README].

That last point has regulatory weight. Because Ackee doesn’t use cookies and doesn’t process personal data in the legal sense, you don’t need a cookie consent banner on GDPR-affected traffic [2]. For a founder running a simple landing page or blog, skipping the cookie banner and the associated JavaScript consent management overhead is genuinely valuable.

The project is built by Tobias Bredenkötter (electerious) and sits at 4,630 GitHub stars [merged profile]. It’s MIT-licensed, which means you can deploy it, fork it, embed it in client projects, or do whatever you want with it without negotiating a commercial license [README].

---

Why people choose it

The core argument for Ackee shows up consistently across the articles: you want traffic data, you don’t want to use Google Analytics, and you don’t want to pay a SaaS subscription for something this simple.

Versus Google Analytics. GA4 is free but it’s Google’s product — every page view you track goes through their infrastructure, feeds their ad targeting, and is subject to whatever terms changes they make unilaterally. For a founder who keeps saying “I should get off Google Analytics,” Ackee is the minimal path to actually doing it. The install looks the same (add a script tag), the dashboard shows the metrics that matter for basic traffic analysis, and you own the data entirely [2][README].

Versus Matomo. Matomo is the most established open-source analytics platform, used by over 1.4 million websites [4]. It offers a closer feature-for-feature replacement for GA, including funnels, A/B testing, heatmaps, and session recordings. But its interface shows its age, many essential features are behind premium plugins costing $229/year each [4], and the self-hosted install involves PHP, MySQL, and a web server — significantly more infrastructure complexity than Ackee’s Docker Compose stack. If you need the power of Matomo, use Matomo. If you just need page views and referrers, Matomo is five times more than you need.

Versus Plausible, Umami, GoatCounter, and the newer privacy-first tools. The self-hosted analytics category has filled up in recent years. Plausible, Umami, GoatCounter, Pirsch, and Simple Analytics all occupy the same “privacy-friendly, simpler than GA” space [5]. Ackee was earlier to this space than most of them. The differentiator is the GraphQL API — Ackee exposes everything through a fully documented API, which means you can query your own analytics data programmatically, build custom dashboards, or feed the data into other systems [README][1]. Umami has a REST API; Plausible has a stats API; but Ackee’s GraphQL surface is explicitly designed as the extension point rather than an afterthought.

Lev Gelfenbuim’s write-up [2] from 2024 puts it plainly: Ackee “demonstrates that it is possible to gain insights into your website’s performance without exploiting the private information of your visitors.” That’s the promise. The reality matches it, with the caveats noted below.

---

Features

What Ackee actually tracks and reports:

Traffic metrics:

• Page views (total and unique) [README]
• Active visitor count in real time [website]
• Average session duration [README]
• Bounce rate (implied through duration and view counts) [README]

Dimension breakdowns:

• Referrers (where traffic came from) [README]
• Language [README]
• Screen dimensions [README]
• Display color depth [README]
• Operating system [README]
• Browser [README]
• Device type [README]

Custom events:

• Track button clicks, newsletter sign-ups, or any custom interaction beyond page views [README][2]
• Events are implemented with the same anonymization constraints as page views [2]

Multi-site support:

• Unlimited domains — add as many sites as you want, all tracked from one Ackee instance [README][website]
• Per-domain tracking codes generated in the settings UI [website]

GraphQL API:

• Everything in the UI is backed by the API — no undocumented internal endpoints [README]
• Fully queryable: you can pull historical views, active visitors, referrers, and statistics for any time interval [README][website]
• There’s a public GraphQL playground at the live demo for poking around [README]

Integrations (third-party):

• ackee-tracker: the official JS tracking library [README]
• ackee-bitbar: macOS menu bar stats widget [README]
• ackee-lighthouse: send Lighthouse performance reports into Ackee [README]
• ackee-report: CLI tool for generating performance reports [README]
• gatsby-plugin-ackee-tracker: Gatsby plugin [README]
• Soapberry: WordPress plugin — notably last updated in 2020 and flagged as potentially unmaintained [3]
• ackee-php: PHP client library [README]
• React hook (use-ackee) mentioned in README

What Ackee does not have:

• Funnels
• Cohort analysis
• User-level profiles or session recordings
• A/B testing
• Heatmaps
• Alerts or anomaly detection
• Team accounts or multi-user RBAC
• Email reports

If any of those are on your requirements list, Ackee isn’t the right tool. It is deliberately minimal.

---

Pricing: SaaS vs self-hosted math

Ackee self-hosted:

• Software: $0 (MIT license) [README]
• Infrastructure: $5–10/mo for a VPS (Hetzner, Contabo, DigitalOcean)
• MongoDB: bundled in the Docker Compose setup or external
• Total: roughly $5–10/mo, unlimited sites, unlimited page views

Elestio managed hosting:

• Starting at $14/mo, includes automated backups, SSL, monitoring, updates [1]
• Good option if you want the data ownership of self-hosted without the operational overhead
• Elestio is SOC2 and GDPR compliant, Trustpilot 4.6/5 from their customer reviews [1]

Competing SaaS tools for context:

• Google Analytics: $0 (you pay with your visitors’ data)
• Matomo Cloud: $19/mo for up to 50,000 monthly hits [4]
• Plausible: $9/mo for up to 10,000 monthly pageviews; $19/mo for 100k
• Simple Analytics: $9/mo
• Fathom: $14/mo

For a founder running 3–5 sites with moderate traffic, self-hosted Ackee at $6/mo on Hetzner beats every SaaS alternative on price. The trade-off is setup time and ongoing maintenance — mostly just keeping Docker updated and not much else.

The managed Elestio option at $14/mo is roughly equivalent to Plausible’s lowest tier, but you own the data and the MongoDB instance rather than having it in someone else’s SaaS [1]. For privacy-sensitive projects, that distinction matters.

---

Deployment reality check

Ackee is a Node.js application backed by MongoDB. The deployment story is more flexible than most self-hosted tools — the README lists 12 deployment paths: Docker Compose, Docker, Helm, plain Node, Netlify, Vercel, Heroku, Qovery, Render, Railway, Koyeb, and Zeabur [README].

For the typical self-hoster (Docker on a VPS):

• You need Docker and docker-compose installed
• Ackee’s docker-compose includes both the Node app and MongoDB
• You’ll want a reverse proxy (Caddy is the easiest) for SSL termination
• CORS headers need to be configured correctly for the tracking script to fire from your actual domains — this is the most common setup stumble [README]

For the serverless path (Vercel or Netlify):

• Lev Gelfenbuim’s guide [2] walks through deploying on Vercel — notably Vercel’s serverless functions mean the MongoDB connection needs to be external (MongoDB Atlas free tier works)
• Vercel deployment means no persistent server, but also means you pay $0 for the compute

CORS configuration matters more than you’d expect. Ackee has explicit documentation on CORS headers because the tracking script runs on your site’s domain and calls back to your Ackee instance’s domain — standard cross-origin behavior. If you misconfigure CORS, tracking silently fails and you’ll spend time debugging missing data [README]. The docs cover it, but this is a common first-time friction point.

What can go sideways:

• MongoDB on a $5 VPS with heavy traffic can get memory-constrained. Budget at least 2GB RAM for a lightly loaded instance; 4GB if you’re tracking high-traffic sites.
• The Soapberry WordPress plugin [3] — the main WordPress integration — hasn’t been maintained since 2020 and has only 10+ active installs. It works for basic use but isn’t production-reliable for a high-traffic WordPress site.
• The project’s GitHub activity suggests it’s in maintenance mode rather than active development. New contributors and issues exist, but major feature additions have slowed. If you’re betting on the tool getting funnels in six months, don’t.

Realistic time estimates:

• Docker Compose on a fresh VPS with a working CORS config: 1–2 hours for someone comfortable with Linux
• Vercel + MongoDB Atlas deployment following Lev’s guide [2]: 2–3 hours
• WordPress integration via Soapberry: 30 minutes, but test it thoroughly [3]

---

Pros / Cons

Pros

• Genuinely cookie-free. Not “cookieless mode with extra configuration.” No cookies, period. GDPR-compliant out of the box without a consent banner [2][README]. For founders tired of cookie consent UX, this is the cleanest path.
• MIT license. You can deploy it for clients, embed it in your product, resell it as a service — no commercial license conversation needed [README].
• Unlimited domains, one instance. One Ackee install tracks all your sites. No per-domain pricing tier [README][website].
• GraphQL API as a first-class feature. Everything the UI shows is API-accessible. You can pull your own traffic data into dashboards, Notion, Slack notifications, or anything else [README][1].
• Flexible deployment. Docker, serverless, Helm, managed PaaS — more options than most self-hosted analytics tools [README].
• Minimal interface that doesn’t get in the way. The UI shows what you need: views, referrers, durations, device breakdown. No 47-tab dashboard to navigate before you find your page view count [website][2].
• Event tracking. Custom events beyond page views are supported and privacy-compliant [2][README].
• Managed option available. Elestio offers fully managed Ackee at $14/mo if you want ownership without operations [1].

Cons

• No funnels, cohorts, session recordings, or A/B testing. If you’re running a SaaS and need to track where users drop off in your onboarding flow, Ackee can’t help. It’s a traffic counter, not a product analytics platform [README].
• Maintenance pace has slowed. The project is maintained by a solo developer. Feature velocity is low. Third-party integrations like the WordPress plugin [3] are years out of date.
• MongoDB is the only database option. Most lightweight self-hosted tools have moved to SQLite or PostgreSQL, which are simpler to operate. Ackee requires MongoDB, which adds operational complexity and memory overhead.
• CORS setup is a foot-gun. If you get CORS headers wrong — which is easy to do when tracking across multiple subdomains — tracking silently fails. The docs explain it, but it’s not obvious for first-time self-hosters [README].
• Small integration ecosystem. A handful of third-party integrations exist; most are maintained by individuals and some are stale [README][3].
• No team features. Single-user auth, no RBAC, no shared access controls. If multiple people need to see the analytics dashboard, they share one set of credentials.
• No alerting. Traffic spike, crawl spike, traffic drop — Ackee doesn’t notify you. Monitoring is entirely external.

---

Who should use this / who shouldn’t

Use Ackee if:

• You’re a developer or technical founder who wants traffic data without touching Google Analytics.
• Privacy compliance is a real concern and you want cookie-free analytics without configuring a complex tool for it.
• You have multiple sites and don’t want to pay per-domain SaaS fees.
• You want to query your analytics data programmatically via GraphQL.
• You’re comfortable with Docker and a basic Linux VPS, or willing to use Elestio’s managed service.

Skip it (use Plausible or Umami instead) if:

• You want simpler setup and more active development than Ackee currently offers.
• You need a maintained WordPress plugin or clean CMS integrations.
• You want email reports or traffic alerts without building them yourself.

Skip it (use Matomo instead) if:

• You need funnels, A/B testing, heatmaps, or session recordings.
• You’re replacing Google Analytics for a serious ecommerce site and need the full conversion tracking stack.
• You need CNIL-approved consent-exempt analytics (Matomo has specific configuration for this; Ackee doesn’t address it [4]).

Skip it (use PostHog instead) if:

• You’re tracking a SaaS product and need user-level events, feature flags, and product analytics alongside web traffic.

Skip it (stay on Google Analytics) if:

• You’re not paying for SaaS analytics and don’t care about the data-sharing trade-off.
• Your reporting relies on GA-specific features (Looker Studio integration, cross-device tracking, Attribution).

---

Alternatives worth considering

From the analytics tool landscape [5] and the review sources:

• Plausible — Closest like-for-like alternative to Ackee. Also cookie-free, simpler install (SQLite, no MongoDB), more actively maintained, SaaS from $9/mo or self-hostable. Better choice if you want something more polished with less maintenance.
• Umami — Similar positioning to Plausible. MySQL/PostgreSQL backend, clean UI, self-hostable. More active development than Ackee.
• GoatCounter — Extremely lightweight, cookie-free, open source. Even simpler than Ackee if you just want page counts and referrers. Less ecosystem.
• Matomo — The full-featured open-source GA replacement. More powerful, more complex to deploy, more costly in infrastructure, but covers everything Ackee doesn’t [4].
• PostHog — Open-source product analytics (funnels, session recordings, feature flags). For SaaS products rather than content sites. Self-hostable.
• Simple Analytics — SaaS-only, no self-host. Cookie-free, clean. If you want zero ops, $9/mo starting.

For a non-technical founder who just wants “not Google Analytics, clean dashboard, privacy-friendly,” the realistic choice is Ackee vs Plausible vs Umami. Ackee wins if you have multiple sites and want the GraphQL API. Plausible wins if you want a more actively maintained project with better UX polish. Umami splits the difference.

---

Bottom line

Ackee does one thing: it counts your web traffic without surveilling your visitors. It does that thing well, it’s MIT-licensed, it deploys via Docker, and it costs you the price of a VPS. If that’s the problem you’re trying to solve — escape Google Analytics, stay GDPR-compliant without a cookie banner, own your data — Ackee delivers. The caveats are real: MongoDB adds operational weight compared to newer tools, third-party integrations are sparse and aging, and the project’s development pace is slow. Newer tools like Plausible and Umami have caught up in features and are more actively maintained. But for a founder who wants a programmatic GraphQL API over their analytics data, or who’s been burned by per-domain SaaS pricing, Ackee remains the shortest path from “I hate Google Analytics” to “I own my traffic data.”

If the setup is the blocker, that’s what upready.dev handles. One-time deployment, your infrastructure, done.

---

Sources

1. Elestio — Managed Ackee as a Service (fully managed hosting from $14/mo). https://elest.io/open-source/ackee
2. Lev Gelfenbuim, lev.engineer — “Beyond Google Analytics: Embrace Privacy with Ackee on Vercel” (Feb 16, 2024). https://lev.engineer/blog/beyond-google-analytics-embrace-privacy-with-ackee-on-vercel
3. ChoosePlugin.com — “Soapberry — WordPress plugin for Ackee analytics”. https://chooseplugin.com/plugin-info/soapberry
4. OpenPanel — “Best Matomo Alternative 2026 — Open Source & Free” (comparison page covering analytics tool landscape). https://openpanel.dev/compare/matomo-alternative
5. W3Techs — “133 Navegg and Ackee alternatives — Traffic Analysis Tools comparison”. https://w3techs.com/technologies/compselect/ta-ackee,ta-navegg

Primary sources:

• GitHub repository and README: https://github.com/electerious/ackee (4,630 stars, MIT license)
• Official website: https://ackee.electerious.com
• Live demo: https://demo.ackee.electerious.com
• GraphQL Playground: https://demo.ackee.electerious.com/api