Cube.dev

Analytics and dashboards directly in your application – with consistent metrics for people and AI agents.

dectria builds analytics and dashboards directly into applications with Cube: metrics are defined once, centrally, and are then available to dashboards, reports and AI agents with the same values and access rules.

NetCero, an ESG platform in which dectria holds a stake, uses Cube for its analytics.

We build the dashboards with the same stack as the rest of your application – in your design, without a separate BI tool that users have to open first.

Official website

Your contact Michael Jauk Contact for AI & data

What is Cube – and when is it worth it?

Cube is a semantic layer between the database and the application. Metrics, dimensions and joins are defined once in the data model and delivered via SQL, REST or GraphQL to dashboards, applications and AI agents – with the same access rules for all. Cube Core is open source; Cube Cloud is the managed option.

Cube is worth it when analytics are part of your product, for example in a SaaS application with many customers, or when several tools need the same metrics. It is not worth it for internal analysis by a few analysts: a BI tool such as Power BI is set up faster there.

Cube or a classic BI tool?

	Cube with custom dashboards	Classic BI tool, e.g. Power BI
Where users see the data	in your application, in your design	in the BI tool or as an embedded report
Typical users	customers and employees in your product	internal analysts
Access rules	central in the semantic layer, the same for every access	in the BI tool
Development effort	higher: dashboards are developed	lower: reports by clicking
Suited for	embedded analytics in SaaS products	internal analysis

What We Build with Cube.dev

Semantic Data Model (Cubes / Measures / Dimensions) Pre-Aggregations & Caching Access Rules & Multi-Tenancy SQL / REST & GraphQL APIs Connecting BI Tools via the SQL API Data Sources such as PostgreSQL / BigQuery / Snowflake Dashboard Development Charts with ECharts / Recharts / D3 Metrics for AI Agents Cube Core & Cube Cloud

Use Cases

Typical Use Cases

Metric Dashboards for ESG and Controlling

Interactive analysis by period, location or category, where every number rests on the same central definition.

Embedded Analytics for SaaS Products

Dashboards directly in your application, in your design, where each customer sees only their own data.

Metrics for AI Agents

Agents query metrics through the semantic layer instead of the database directly – with the same definitions and permissions as the dashboard.

References

Selected Projects

Web Application

Klanglicht - Art Festival of Bühnen Graz

Klanglicht Audio Guide

Mobile App

Radio Soundportal

Soundportal App

Web Application

Österreichische Raiffeisen-Sicherungseinrichtung eGen (ÖRS)

Deposit Insurance Payout Platform

View all references

FAQ

Cube.dev FAQ

Why Cube instead of a classic BI tool?

Because analytics in your product work differently from internal reports. BI tools bring their own interfaces that hardly fit into your application. Cube provides the query engine without an interface, and we build the dashboards in your application's design. For internal analysis, a BI tool often remains the better choice.

How fast is Cube with large data volumes?

Cube pre-computes frequently queried metrics with pre-aggregations and refreshes them in the background. Dashboards then read from these pre-computed tables instead of evaluating all raw data each time. We decide which metrics to pre-compute based on the actual queries.

Can Cube run multi-tenant?

Yes. Access rules live in the semantic layer and apply to every query – whether it comes from a dashboard, an API or an AI agent. This way each customer sees only their own data, without every dashboard needing its own filters.

Can AI agents access metrics through Cube?

Yes. Agents query through the semantic layer instead of the database directly. They use the same metric definitions as your dashboards and are subject to the same access rules. This prevents an agent from calculating a metric differently from the report.

Every project starts with a conversation.

Let us talk about your individual needs and goals.

Start a project