Chartbrew

Open-source analytics and reporting, honestly reviewed. Not what’s on the marketing page — what you get when you actually self-host it.

TL;DR

• What it is: Open-source (FSL-1.1-MIT licensed) reporting platform that connects to databases and APIs to build live dashboards, embeddable charts, and automated client reports [3][README].
• Who it’s for: Agencies, freelancers, and small teams who need to deliver branded dashboards to multiple clients without paying per-seat or per-dashboard fees. Also technical founders who want a self-hosted Metabase alternative with a client-access layer [1][3].
• Cost savings: Chartbrew Cloud starts at $29/mo for 10 dashboards and clients. Self-hosted on a $6–10 VPS eliminates that entirely, with unlimited dashboards and clients [1].
• Key strength: Multi-tenant client reporting with white-labeling and granular permissions — a feature set that Metabase buries behind its Pro tier [1][2].
• Key weakness: The v4 license changed from MIT to FSL-1.1-MIT, which restricts building competing commercial services on top of it. Not fully open source in the OSI sense. GitHub star count (3,681) is modest compared to Metabase (38K+), signaling a smaller community and ecosystem [3][README].

---

What is Chartbrew

Chartbrew is a self-hosted web application that connects to your databases and APIs, lets you build charts and dashboards in a visual editor, and then share or embed those dashboards with team members or external clients. The project started as a straightforward open-source Metabase alternative, but it has drifted into a clearer niche: client-facing reporting for agencies and small teams [1][README].

The homepage headline — “Actionable insights for your team and clients” — is more honest than most. The product’s differentiator isn’t raw query power (Metabase beats it there) or SQL-first workflows (Evidence.dev does that better). It’s the client management layer: you can create separate client accounts with scoped access, replicate dashboards across clients using templates, schedule automated snapshots delivered by email or Slack, and white-label the whole thing [1][3].

The GitHub README describes it as an “open-source reporting platform to build and share live dashboards from APIs, SQL and NoSQL databases, with powerful AI assistant, scheduling, and embeddable charts” [README]. That’s accurate and covers the main use cases without overselling.

As of this review, the project has 3,681 GitHub stars. That’s honest to include: it’s a healthy indie project, not a category leader. Metabase has over 38K stars, Redash has around 26K. Chartbrew is actively maintained — v4 launched April 2025 with significant changes including a license update, new dashboard filters, and scheduled snapshots [3].

One thing to flag upfront: Chartbrew v4 changed its license to the Functional Source License 1.1 (FSL-1.1-MIT). This allows you to use, modify, and embed Chartbrew in your own applications and SaaS platforms. What it restricts is building a competing hosted analytics service on top of it [3]. For most self-hosters and agencies, this doesn’t matter at all. If you were planning to resell a Chartbrew-based analytics product to hundreds of clients as your core business, read the license carefully.

---

Why people choose it

SaaSHub’s review page [1] and the alternatives comparison [2] both position Chartbrew squarely against Metabase, Redash, and Databox. The choice pattern that emerges is consistent: people pick Chartbrew when they need client-facing reporting with access controls, and they pick Metabase when they need internal BI with complex SQL.

Versus Metabase. Metabase is the obvious comparison. It’s more powerful for internal analytics — better SQL editor, questions and collections, stronger embedding API, and a much larger community. But multi-tenant client access requires Metabase Pro at $500/mo for 5 users [2]. Chartbrew’s $29/mo SaaS or self-hosted plan gives you granular client permissions, white-labeling, and template-based dashboard replication for multiple clients out of the box [1][3].

One testimonial on the Chartbrew homepage puts the positioning clearly: “Chartbrew has helped us move away from having to constantly update clunky Google-based charts, but what most impresses me is the responsiveness and the helpfulness of the people behind Chartbrew.” — Schuyler, Fairchain [homepage]. That’s a client-side reporting person, not a data engineer.

Versus Redash. Redash is heavily SQL-focused and designed for internal analyst teams. Chartbrew has a more consumer-facing visual editor and the client management features Redash lacks entirely [2].

Versus Databox. Databox is SaaS-only, starts at $159/mo for the Professional plan, and charges per user [2]. Chartbrew self-hosted eliminates that recurring cost entirely. The trade-off is you manage the infrastructure.

The AI angle. Chartbrew v4 includes a schema-aware AI assistant that can answer natural language questions about your data and help write SQL queries [3][website]. The website pitches it as “Data Conversations in Slack” — you mention the bot in a Slack channel and ask questions about your metrics [5][website]. Whether this is genuinely useful or demo-ware depends on your data quality. The query-writing assistant for SQL databases is a real productivity feature for non-SQL users [3]. The autonomous “ask questions” pitch is marketing.

---

Features

Based on the README, v4 release notes, and the website:

Core reporting engine:

• Visual chart builder (line, bar, doughnut, KPI, table, and more) [README]
• Editable dashboards with drag-and-drop layout [README]
• Compact mode for dense, cleaner dashboard layouts added in v4 [3]
• Auto-update schedules — dashboards refresh on your set interval [1]
• Fully rebuilt dashboard filters in v4: custom filter sets, date ranges, variables like start_date and end_date in queries, shared or browser-local [3]
• Public data filters — viewers can filter without logging in [1]

Data sources:

• MySQL, PostgreSQL, MongoDB, Firestore, Firebase Realtime Database [1][README]
• Custom REST API connections [1]
• Google Analytics, Customer.io [1]
• Amazon RDS, TimescaleDB [1]

Client and team features:

• Client accounts with scoped access — clients only see what you give them [1][3]
• Granular permissions per team member and per client [1]
• Dashboard templates — replicate a dashboard structure across multiple clients [1][3]
• White-label reports with custom branding [1]
• Scheduled snapshots delivered via email or Slack — no client login required [3]
• Excel and PDF exports [1]

Sharing and embedding:

• Embeddable charts via iframe or API [1][3]
• Shareable public report links [1]
• Multi-tenant support for isolating clients [1]

AI features:

• Schema-aware AI assistant for query help and natural language data questions [3][website]
• Slack bot integration: ask data questions directly in Slack channels [5][website]
• AI-powered query assistant for SQL databases [3]

Deployment and infrastructure:

• Docker image with Docker Compose setup [README]
• DigitalOcean 1-click marketplace deployment [README]
• NodeJS 22 + MySQL or PostgreSQL + Redis required [README][3]
• CircleCI and GitHub Actions CI pipelines [README]

---

Pricing: SaaS vs self-hosted math

Chartbrew Cloud:

• Free trial: 14 days without a credit card; 30 days if you add a payment method [3]
• Starter: $29/mo — 10 dashboards and clients, unlimited connections and charts [1]
• Higher tiers: data not available in reviewed sources — contact sales or check chartbrew.com/pricing for current numbers

Self-hosted:

• Software license: $0 (FSL-1.1-MIT) [3]
• VPS to run it: $6–10/mo on Hetzner or DigitalOcean
• Your time to set it up

Metabase for comparison:

• Open Source Edition: free, self-hosted, limited embedding and no SSO
• Pro: $500/mo for 5 users — where multi-tenant and advanced embedding live
• Enterprise: custom pricing

Concrete math for a typical agency:

Say you’re an agency managing dashboards for 8 clients. On Metabase Pro that’s $500/mo. On Chartbrew Cloud at current pricing, data isn’t available for tiers above 10 clients — but self-hosted Chartbrew handles unlimited clients for the cost of a VPS: roughly $8/mo on Hetzner.

Over a year: Metabase Pro ≈ $6,000. Chartbrew self-hosted ≈ $96 + setup time.

For a freelancer or 2-person agency, this math is not subtle. The gap closes if you’re comparing Chartbrew against Metabase’s free open-source tier (which you can also self-host), but the free Metabase edition doesn’t include the client access controls that are Chartbrew’s whole differentiator.

---

Deployment reality check

The README is honest about prerequisites: Node 22, MySQL 5+ or PostgreSQL 12.5+, Redis 6+ [README]. The Docker path is the recommended route, and the docker-compose setup is reasonably straightforward. DigitalOcean marketplace has a 1-click droplet [README], which is the path of least resistance for non-technical users.

What you actually need:

• A Linux VPS with at least 2GB RAM
• Docker and docker-compose
• A running MySQL or PostgreSQL instance (can be on the same server or external)
• Redis (can be bundled)
• Domain + reverse proxy (Caddy or nginx) for HTTPS
• A 32-byte AES encryption key for CB_ENCRYPTION_KEY — generate with one Node command [README]

The Slack integration adds complexity. Setting up the Slack bot for self-hosted requires creating a Slack app, updating manifest URLs in four separate places (slash commands, OAuth redirect, event subscriptions, interactivity), pulling three credential values, adding them to your .env, and restarting the server plus rebuilding the client [5]. It’s doable but not five minutes — budget an hour if you’ve done it before, two if you haven’t.

The v4 breaking changes matter if you’re upgrading. The move to Node 22 as the minimum and the upgrade to Express v5 mean you can’t just pull the latest image if you’ve made custom code changes — there will be conflicts [3]. For vanilla deployments with no modifications, a pull-and-restart should work.

What the documentation doesn’t make obvious: you need the database running and empty before you start the Docker container. The setup sequence matters — getting this wrong is the most common source of frustrating first-run failures.

Realistic time estimate for a technical user who’s run Docker before: 45–90 minutes to a working instance including domain setup. For a non-technical founder following the DigitalOcean 1-click path: 2–3 hours including DNS propagation wait time. If you’ve never managed a server, factor in either a full day or hiring someone to do it once.

---

Pros and Cons

Pros

• Client management built-in. Scoped client access, template-based replication across clients, scheduled snapshot delivery — this is the feature set agencies pay $500/mo for in Metabase [1][3].
• Embeddable charts via iframe or API. Building analytics into your own product takes hours, not weeks [1][website].
• Multiple database types in one tool. MySQL, PostgreSQL, MongoDB, Firestore, REST APIs — you’re not forced to migrate data to query it [1][README].
• Scheduled report delivery. Clients get automated email/Slack snapshots without needing to log in [3]. This is genuinely useful for non-technical clients.
• DigitalOcean 1-click deployment. The lowest-friction path to self-hosting in this category [README].
• Active development. v4 shipped in April 2025 with substantive changes, not just a version bump [3]. 468 code changes since v3 [3].
• Free trial without a credit card. 14 days on the cloud version to evaluate before committing [3].

Cons

• FSL-1.1-MIT is not MIT. The v4 license change restricts building competing hosted analytics services. If you were counting on MIT for commercial redistribution, read the FSL before committing [3]. For personal self-hosting and agency use, it doesn’t matter.
• Smaller community than the alternatives. 3,681 stars vs. Metabase’s 38K+ and Redash’s 26K. Fewer tutorials, fewer integrations, fewer community answers when things break [README][2].
• Requires external database and Redis. Not a one-container deployment like some tools. The setup dependency chain (Chartbrew + MySQL/PostgreSQL + Redis) is manageable but not trivial [README].
• The Slack integration setup is involved. Four manifest URL replacements, three credentials, env var changes, server restart, client rebuild [5]. Fine for a sysadmin, annoying for everyone else.
• AI features are limited by your data quality. The “ask anything about your data” pitch assumes clean, queryable data. The SQL query assistant is real. The conversational analytics angle is promising but not turnkey [3][website].
• Limited third-party review coverage. The reviews available are largely from SaaSHub aggregation and the vendor’s own blog [1][2][3]. There’s no deep independent teardown from a major tech publication, which makes independent validation of claims harder.
• Pricing tiers above Starter not documented in available sources. For larger deployments, you’d need to check current pricing directly — what’s available here covers the Starter tier only [1].

---

Who should use this / who shouldn’t

Use Chartbrew if:

• You’re an agency or freelancer managing dashboards for multiple clients and you need client accounts with scoped access.
• You’re currently handing clients Google Data Studio dashboards and want something cleaner and more professional.
• You want to embed live charts into your own product or client portal without building an analytics stack from scratch.
• You’re comfortable with Docker deployment, or willing to use the DigitalOcean 1-click path.
• Your data sits in MySQL, PostgreSQL, MongoDB, or is accessible via REST API.

Skip it (use Metabase instead) if:

• You need internal BI for your own analytics team, not client-facing reporting.
• You need advanced SQL tooling — named queries, saved questions, collections, SQL variables.
• You want the largest open-source BI community and the most tutorials when you get stuck.
• The FSL license restriction matters for your use case (e.g., building an analytics product to resell at scale).

Skip it (use Evidence.dev instead) if:

• Your team is engineer-led and wants version-controlled, SQL + Markdown reports that deploy like a static site.
• You value reproducibility and git history over a visual drag-and-drop editor.

Skip it (use Grafana instead) if:

• Your primary data source is time-series metrics, logs, or infrastructure monitoring.
• You need a battle-tested tool with a massive plugin ecosystem for operational dashboards.

Stay on a SaaS tool if:

• You have no one who can manage a Linux server, and the DigitalOcean 1-click path still feels out of reach.
• Your compliance requirements prohibit self-hosted infrastructure.
• You need a vendor with an enterprise support contract and SLA.

---

Alternatives worth considering

From the SaaSHub alternatives list [2] and the competitive landscape:

• Metabase — the most direct alternative for internal BI. Larger community, more powerful SQL tooling, multi-tenant client access requires Pro ($500/mo). Self-hosted community edition is free. [2]
• Redash — SQL-first, internal-team-focused, open source. No client management layer. Good for analyst teams, not agencies. [2]
• Evidence.dev — code-first SQL + Markdown reporting, version-controlled, deploys as a static site. Excellent for engineers who want reproducible reports, not visual builders. [2]
• Grafana — time-series and infrastructure monitoring. Different use case entirely, but people sometimes consider it for business dashboards. Better for ops than for client reporting.
• Databox — SaaS-only, starts at $159/mo, polished interface, good social/marketing data integrations. If self-hosting is off the table and budget allows. [2]
• Explo — SaaS, focused on embedded analytics. More mature embedding story than Chartbrew for product teams building analytics into their apps. [2]
• Superset — Apache-licensed, SQL-heavy, powerful for large data teams. Steeper setup than Chartbrew. No client management features.

For an agency building client dashboards, the realistic shortlist is Chartbrew vs Metabase. Pick Chartbrew if client accounts and multi-tenant templates are your core need. Pick Metabase if SQL power and community depth matter more than client management.

---

Bottom line

Chartbrew has found a defensible niche that the bigger BI tools don’t cover cleanly: client-facing reporting for agencies and small teams who need to deliver branded dashboards to multiple clients without paying enterprise prices. The core feature set — client accounts, template replication, scheduled snapshot delivery, embeddable charts — genuinely covers a workflow that otherwise costs hundreds of dollars a month in Metabase or Databox. The self-hosted path on a $8 VPS makes the cost argument obvious.

The honest caveats: the FSL-1.1-MIT license in v4 is not fully open source, the community is small, and this is largely a one-person project from maintainer Razvan Ilin. That last point cuts both ways — the testimonials about responsive support are credible, but betting your agency’s client reporting stack on a solo-maintained project is a real risk to weigh. If you can accept that trade-off (or contribute to the project yourself), Chartbrew is one of the more focused and honest tools in the self-hosted analytics space.

---

Sources

1. SaaSHub — Chartbrew Reviews and Details (features, pricing, integrations). https://www.saashub.com/chartbrew
2. SaaSHub — Chartbrew Alternatives & Competitors (competitive landscape, alternatives list). https://www.saashub.com/chartbrew-alternatives
3. Razvan Ilin, Chartbrew Blog — “Chartbrew v4 - what’s new” (v4 features, license change, deployment changes, April 27, 2025). https://chartbrew.com/blog/chartbrew-v4-whats-new/
4. Razvan Ilin, Chartbrew Blog — “Create your Strapi visualization dashboard with Chartbrew” (integration tutorial, REST API connection setup). https://chartbrew.com/blog/create-your-strapi-visualization-dashboard-with-chartbrew/
5. Chartbrew Docs — Slack Integration (Slack bot setup, self-hosted configuration, prerequisites). https://docs.chartbrew.com/integrations/slack

Primary sources:

• GitHub repository and README: https://github.com/chartbrew/chartbrew (3,681 stars, FSL-1.1-MIT license)
• Official website: https://chartbrew.com
• Documentation: https://docs.chartbrew.com