Types of Metrics

There are four kinds of metrics you can add to a model: Additive metrics, non-additive metrics, semi-additive metrics, and calculations. This section describes each type of metric and how AtScale's aggregate system manages them.

You add a metric to a model by first identifying the fact dataset of your model, then applying aggregate calculations to the quantifiable dataset columns. The type of aggregate calculation you apply determines the type of metric it is.

📄️ Additive Metrics

Additive metrics are those whose values can be summarized for any dimension attribute of the model, and the results can be combined consistently.

📄️ Non-Additive Metrics

Non-additive metrics are those that cannot be summed across any dimensional groupings using basic addition, since this would usually produce an inaccurate result. The most common example of a non-additive metric is a distinct count of an attribute value.

📄️ Semi-Additive Metrics

Semi-additive metrics are those whose values can be summarized for some dimensions of a model, but not all. Ratios such as Average are also considered semi-additive metrics.

📄️ Calculations

A calculation uses an MDX or DAX expression, often in the form of a mathematical formula, to combine, evaluate, or manipulate other metrics defined in the model.