Datasette

What is Datasette?

Datasette is an open source tool for exploring and publishing data. It takes a SQLite database and instantly serves it as an interactive website — with faceted search, filtering, sorting, and custom SQL queries — alongside a matching JSON API, all from a single command. Built in Python, it’s designed to make structured data easy to share.

What is Datasette best for?

Datasette is best for publishing a fixed dataset for people to explore. Data journalists, researchers, archivists, museums, and local governments use it to put open data online without building a backend. It shines when your data is essentially read-only and you want a shareable, queryable interface fast.

What can Datasette do?

• Serve any SQLite database as a browsable website with one command (datasette serve data.db)
• Auto-generate a JSON API for every table and query — no backend code required
• Run arbitrary read-only SQL from the browser, with a query timeout (default 1000ms) as a safety valve
• Facet, filter, sort, and full-text search across tables
• Publish straight to Google Cloud Run, Heroku, or Vercel with datasette publish
• Extend it with 150+ plugins for charts, maps, authentication, and data import
• Run entirely in the browser via Datasette Lite (WebAssembly), with no server at all

Where does Datasette fall short?

• It’s built for read-only publishing, not live transactional apps. Databases are opened immutable by default, so it’s the wrong tool if you need constant writes or multi-user editing (write features exist via plugins but aren’t the core use case).
• It’s SQLite-only. There’s no native PostgreSQL or MySQL backend, so getting relational data in usually means an ETL step to convert it to SQLite first.
• Version 1.0 has spent a long stretch in alpha, so some newer APIs and write-oriented features are still stabilizing.

Is Datasette free?

Yes — Datasette is fully open source under the Apache 2.0 license and free to self-host; you only pay for wherever you deploy it. There’s also Datasette Cloud, a paid managed service that hosts your instances and adds CSV/SQLite upload with no setup, for teams that don’t want to run it themselves.

What does Datasette replace?

Datasette is a lightweight, self-hosted alternative to business intelligence and data-sharing products like Tableau, Looker, and Airtable. It won’t match their dashboarding or spreadsheet-style editing, but for the specific job of publishing a dataset as an explorable website and API, it’s far simpler and free.

FAQ

Is Datasette open source? Yes. Datasette is released under the Apache 2.0 license — a permissive, OSI-approved open source license — and developed in the open on GitHub.

Can I self-host Datasette for free? Yes. The software is free; you install it with pip, pipx, Homebrew, or Docker and run it on your own server, paying only for the hosting. Datasette Cloud is the optional paid managed version.

Is Datasette a good Tableau alternative? For publishing and exploring a dataset, yes — it’s simpler and free. For rich interactive dashboards and visual analytics, Tableau and Looker still do far more out of the box.

What do I need to run Datasette? Python 3.8 or newer and a SQLite database. A basic instance is one command; production deployments typically use Docker or a publish target like Cloud Run.