Turn data into decisions
Sovereign by design. Built in Utrecht. From ingestion to insights, all in one place. On infrastructure you actually control.
Dashboards
Live views of the metrics that matter. The whole team works with the same trusted numbers, no SQL required.
Notebooks
When something looks off in the data, open a notebook and trace it back to the source. In SQL or Python, reproducible every time.
Agents
Set agents to watch your data and act the moment something changes. Answers and alerts, without the manual check.
Observe
One number looks off. Everyone opens the same dashboard, sees the same spike, and knows what to do next. No exports, SQL or tickets to the data team.
More than a chart on a wall.
Dig deeper yourself
Click any number to break it down by cohort, region, or time. No new query or analyst needed.
Share by link
Send a live view to anyone on the team. They see what you see, with the right access rules already applied.
Always current
Every chart reads directly from the lakehouse. The values on screen are always up to date.
Powered by Apache Superset
Read more about the platform stack
Explore
To investigate, an analyst opens the churn investigation in a notebook, exploring the cohort in SQL or Python, right next to the data.
Go as deep as the question needs.
SQL or Python
Query in SQL, model in Python, and mix both in a single reactive notebook.
Reproducible by default
Marimo stores notebooks as plain Python in git. No hidden state, no JSON diffs.
Share the finding
Publish a notebook as a report so the whole team sees exactly how you got there.
Powered by JupyterHub + Marimo
Read more about the platform stack
Automate
Set an agent to watch any metric. When something shifts, it alerts the team drafting a summary. All within the access rules you already set.
Agents are in active development. Want early access? Talk to us.
From answer to action.
Ask in plain language
Query your data conversationally. The agent will read the catalog and write the SQL for you.
Acts on a change
Set it to watch a metric and trigger an alert, a summary, or a downstream workflow.
Same rules as your team
Agents will run under the same Zitadel policies, never more access than the person asking.
Powered by MCP
Read more about the platform stack