Metrics

Last updatedUpdated: Mar 13, 2026 by Simon Späti · CreatedCreated: Sep 19, 2022

Metrics, KPIs (Key Performance Indicators), and Calculated Measures are terms forming the foundation of how business performance is measured and defined.

Understanding and defining an organization’s KPIs is essential, requiring a shared comprehension across the team. Metrics, central to Metrics Layers and Semantic Layers, are commonly displayed in business reports and dashboards, offering organization-wide accessibility.

Definition of a Metric/Measure

A metric or measure is a context-sensitive expression that can be evaluated across various dimensions to produce a meaningful value, typically representing a key business indicator or performance measure, without requiring explicit SQL joins or aggregations for each calculation.

A Metric, synonymous with KPI or a calculated measure and outlining business performance. They typically emerge in business reports and dashboards, providing a clear view of organizational performance to all members.

Consider operational metrics that reflect your company’s performance and service levels, or financial metrics that illustrate its fiscal health. These metrics are often defined through elaborate SQL statements within BI Tools.

Calculated measures, integral to metrics, relate specifically to Dimensions, traditionally mapped in a Bus Matrix. Dimensions represent categorical aspects used for segmentation, filtering, grouping, slicing, and dicing. Examples include sales amount, region, city, product, color, and distribution channel. These concepts are further explored in Dimensional Modeling.

Metrics are the best medium for data modeling

I’d always advocate that measures and dimensions are the key and the medium in which we model everything. Even in Excel, but usually people just have a data set and want to pivot.

I always try to support and bring a good data model and metrics design as a priority.

# Different Metrics Types

MetricFlow defines metrics as

Metrics, which is a key concept, are functions that combine measures, constraints, or other mathematical functions to define new quantitative indicators. About MetricFlow | dbt Developer Hub

And support different metric types:

Conversion: Tracks when a base event and a subsequent conversion event occurs for an entity within a set time period.
Cumulative: Aggregates a measure over a given window.
Derived: Defines a metric as an expression of other metrics, which allows you to do calculations on top of metrics.
Ratio: Defines a metric as the ratio of two measures, such as revenue per customer.
Simple: Defines a metric that directly references a single measure.