DataxLand documentation

Get started with DataxLand — connect data, review Si's proposal, read verified dashboards and ask Si questions.

DataxLand is the AI data team for companies without one: it connects any business data (Excel, databases and 500+ sources) for any industry and team, builds verified dashboards on its own, and lets you ask Si, the AI data assistant. Si works out what your data means and asks when something is unclear; you approve before anything is built.

Where to start

You want toPage
See what DataxLand does in five minutes, then get your first dashboardQuickstart
Understand workspaces, their states and rolesWorkspace
Connect Excel, a database, an API, a marketplace or a connectorConnect data
Review how Si understood your data before buildingReview and build
Ask Si in the appAsk Si in the app
Know exactly how a number is calculatedMetrics and the semantic layer
Get reports and alerts, or ask Si in Slack, Teams or TelegramReports and alerts
Connect Claude, Cursor or your own agentMCP server
Invite colleagues and turn on 2FAMembers and roles
Set up SSO, custom roles and row-level accessSSO · Roles · Row-level access
Embed a dashboard in your own app or portalEmbed a dashboard
Run DataxLand on your own infrastructureSelf-hosting
Check your plan's limitsPlans and quotas

Screen labels are quoted exactly as they appear in the app.

Three data layers

Data flows through three layers: ingest (kept exactly as received), transform (mapped to your industry's standard model) and metric modeling (metrics with a definition and a version). Si proposes how to read and map your data; you approve before anything is built. Details in Architecture.

Something not working?

See FAQ and troubleshooting and the Glossary.

For AI agents

Every docs page has a Markdown version: add .md to the URL (for example /en/docs/mcp.md). The index for agents is /llms.txt; the full text is /llms-full.txt. See Set up with an AI agent.

Markdown