# Redash

> Query, visualise, share — self-hosted and free, with no vendor above it

Redash is a BSD-licensed query-and-dashboard tool: analysts write SQL against any connected source, turn results into visualisations, assemble them into dashboards, and share them — running entirely on the team's own infrastructure, with no commercial edition to upgrade to.

Website: https://redash.io

## Use it when

- The team wants SQL, charts, dashboards, and alerts on its own servers, free, with no commercial tier being sold above it.
- Source spread is wide: around fifty connectors including MongoDB, Cassandra, Elasticsearch, Prometheus, and SaaS APIs beside the warehouses.
- The unit of work should stay a saved query: nothing to model before the tool is useful, nothing to maintain to keep it useful.
- Simple threshold alerts on scheduled queries cover the alerting needs.

## Think twice when

- Non-technical users need to author: the builder is a SQL editor with a schema browser.
- Governance needs are real: no row-level security and no column masking; the documentation suggests restricted database users instead.
- You want active feature development: the community project is maintained, with a few releases a year rather than daily motion.

## How it runs

Self-hosted only, as Python services with Postgres and Redis underneath. Queries push down to the source and cache their latest result on a refresh schedule; embedding is a secret link the docs warn against using for anything sensitive. BSD-licensed; the hosted service closed after the Databricks acquisition.

## Details

### General

| Attribute | Value |
| --- | --- |
| License | Open source (permissive) |
| SPDX identifier | BSD-2-Clause |
| Deployment | Self-hosted |
| Workload | Interactive |
| Operational complexity | Medium |
| Pricing | Free |
| Language | Python, JavaScript |
| Repository | https://github.com/getredash/redash |
| Documentation | https://redash.io/help/ |
| Stars | 28.8k |

### Business Intelligence

| Attribute | Value |
| --- | --- |
| Authoring model | A SQL editor with a schema browser |
| Semantic layer | No modelling layer: the unit of work is a single saved query |
| Connectors | Around fifty sources, and the widest spread here, Postgres, MySQL, SQL Server, Oracle, Snowflake, Redshift and BigQuery, but also MongoDB, Cassandra, DynamoDB, Elasticsearch, Prometheus, Google Analytics, Salesforce, Jira, CSV, JSON and a Python source |
| Query execution | Pushdown to the source; Redash caches each query's latest result and refreshes on a schedule |
| Chart library | A table plus twelve types (line, bar, area, pie, scatter, boxplot, cohort, counter, funnel, map, pivot, Sankey, sunburst and word cloud) with no documented way to add your own |
| Definitions in | Redash's own Postgres metadata database |
| Access control | Groups own data sources, and a user gets full access or view-only, no row-level security and no column masking, and the docs tell you to make a restricted database user instead |
| Alerts and delivery | A threshold on one column of a scheduled query's result, evaluated after every run, flipping between OK and TRIGGERED; destinations include email, Slack, Teams, Discord, Mattermost, PagerDuty, Datadog, Asana and a plain webhook |
| Embedding | A secret link or a full-screen URL in an iframe, but the docs warn that anyone can change parameters on an embedded dashboard, so it is not for embedding anything sensitive |
| Billing unit | Free and self-hosted only: the hosted service closed after the Databricks acquisition, and nothing has replaced it |

## Capabilities

- [Business Intelligence](https://matca.io/capabilities/business-intelligence)

---

Source: https://matca.io/redash
Last updated: 2026-08-31T11:40:49.329Z