Cube (semantic layer)

Data platformAnalytics

Universal semantic layer defining metrics once and serving them to every BI consumer.

Updated July 2026 cube.dev

Overview

Cube is a semantic layer that sits between the data warehouse and everything that consumes metrics — BI dashboards, embedded customer- facing analytics, and increasingly AI agents. Teams define measures and dimensions once, in code, and Cube serves consistent numbers through SQL, REST, and GraphQL APIs with caching, pre-aggregation, and access control. In a revenue stack it is how usage and revenue metrics come to mean the same thing in every tool: the definition of billable usage or net revenue lives in one governed place instead of five dashboards.

Capabilities on the RevOps map

Which of the capability map's modules Cube (semantic layer) covers — each links to the module's own page, with every tool that supports it.

Module Phase Depth Note
Fulfill & Bill
SQL-Based Billable Metrics Consume & Meter Core Defines the governed usage metric; rating and invoicing happen in downstream billing systems.
Grow Revenue
Custom Report Builder Platform & Intelligence Partial Powers report builders and embedded analytics via API rather than shipping an end-user reporting UI of its own.

What makes it different

It is consumer-agnostic and headless — not welded to one BI vendor — and its caching layer makes it practical to expose governed metrics inside your own product, not just to internal analysts. That embedded, API-first posture is what separates it from BI-native semantic layers.

Frequently asked questions

Is Cube a BI tool?

No — it is the layer BI tools sit on. Cube holds the metric definitions and answers queries; visualization happens in whatever consumes it, from Looker-style dashboards to charts embedded in your product. The payoff is that every consumer reports the same number.

Can Cube meter usage for billing?

It can define and serve the metric — say, billable API calls per account per month, computed from warehouse events — but it does not rate against a price book or generate invoices. Teams pair a warehouse-plus-semantic-layer metric definition with a billing engine that consumes the aggregate.

Back to stack & tools