Features
Publish from the desktop
Author your reports with the Bistro desktop app. One click pushes the report — data, logic, and layout together — to your private workspace. No separate ETL pipeline, no manual upload steps, no fragile cron job keeping things in sync.
- Data, report logic, and page layouts travel together in a single file — nothing to wire up separately
- Transformations re-applied on every refresh, so the shapes your viewers see match what you authored
- Balances at the end of the period (year-end / quarter-end / month-end) worked out per visual cell, the way Power BI analysts expect
- Native builds for macOS, Windows, and Linux — Linux is a first-class platform, not an afterthought
Bring in data from where it lives
The desktop app reads from:
- Files: CSV, TSV, Parquet, JSON and Excel, or a whole folder of them combined. Files that aren't UTF-8 (older Windows exports, for example) are recognized, and you can pick the encoding yourself
- The web: a file at a web address, or a shared Google Sheet
- Cloud storage: Amazon S3 (and S3-compatible storage), Google Cloud Storage and Azure Blob Storage, plus Delta Lake and Iceberg tables
- Databases: PostgreSQL, MySQL, SQL Server and SQLite built in. Oracle, Snowflake and BigQuery are add-ons the app installs the first time you pick them (Linux today; macOS and Windows coming)
Analysis built in
Common analyses are a few clicks, not a formula hunt:
- Time comparisons: year to date, last year, year-over-year growth, previous period
- Customer patterns: new versus returning, lapsed and reactivated customers, like-for-like stores
- Data recipes: ABC classification, basket analysis (bought together), survey cross-tabs, transitions between tiers, events in progress on each date
- Show value as: % of total, running total, rank, moving average, change from the previous value, right on the visual
- Analyze tab: PCA, clustering, regression, classification, decision trees and outlier detection. They're fitted on your data and applied again on every refresh
- Explore Data: a notebook of charts and SQL over any step of your pipeline, to look around before building a report
- Data tests: checks such as "no blanks", "unique" or your own rule, run on every refresh, so a broken upstream shows up before your viewers see it
Every step is written in BQL, Bistro's readable pipeline language, so you can see and edit exactly what a recipe did.
Role-based access, billing-aware
Per-report roles that match how analytics teams actually work:
- Author — the user who first published. Can rename, delete, share, refresh, and download. Auto-stamped at publish time.
- Contributor — paid seats. Full read + write access. Can publish new reports, refresh data, share with others, annotate.
- Viewer — free seats. Read-only. Can open reports, see notes others have left, see who has access. Can't write or download.
Adding a user as a Contributor automatically updates your subscription; Viewers stay free. The math is simple: you only pay for people who write, never for people who read.
Refresh upstream data on demand
Reports stay live against your source databases:
- PostgreSQL and MySQL refresh on the server today. Other sources show the data as of your last publish; re-publish to update them
- Refresh re-pulls from your database, re-applies your transformations, and updates the report — viewers see the new numbers on next open
- Credentials encrypted at rest, per-user-per-report (no shared service account leaking access)
- Triggered manually by Contributors; never auto-runs without someone asking
Annotate the canvas
Reports get a comment thread, in-context. Drop a pin anywhere on a visual to start a discussion:
- Pins survive republish; the conversation outlives the underlying report version
- Drag to reposition; edit or delete your own; report Authors can moderate
- Viewers see notes others have left — turn the report into a collaborative document, not a static screenshot
Built for sharing across an organization
- Auto-grants — when someone joins your org, they get the right access to every existing report based on their role
- Cross-org access controlled — sharing happens within an org by default; no accidental external leaks
- Audit-friendly — every grant lives in the database with a timestamp; nothing relies on email forwards or share links
Coming soon
- Stripe / Lemon Squeezy partner program for accounting firms
- External-viewer share links (read-only public)
- Slack / email scheduled delivery
- Live refresh on the server for SQL Server, Oracle, Snowflake, BigQuery, cloud storage and web sources (the desktop app already imports them)