Metric vs Measure and Best Practices

  • 14 August 2020
  • 2 replies

What is the difference between Metric and Measures .... Are there best practices around when to use which one ?


Best answer by Pavel 15 August 2020, 16:45

View original

2 replies

Userlevel 3

Good question. Metrics are numerical values that can be either summarizations of facts or individual numerical records and can be built or added under Reports, whereas measures are directly added from the data catalog. That said, it should be useful for you to know that Analytical Designer also allows you to Create Calculated Measures in Insights, but you must keep in mind that they are available only in the insights where you create them since Analytical Designer does not save these calculated measures to the Data Catalog. To use them elsewhere, you need to create the calculated measures again in another insight. If you want to avoid this, it is better to create a custom metric (make sure you click the Add to Global Metrics checkbox.
This makes your metric available for use in any reports/insights within the project.) so you can add it to any of your insights.

Further reading: Create Calculated Measures in InsightsCreating Metrics and Reports


Userlevel 2

For the sake of clarity, let me add a few points if the original question was mostly about the terminology:

  • Technically speaking, on the API level we are distinguishing Facts (numeric columns in raw data) and Metrics (calculated numbers).
  • The responsive dashboards and Analytical Designer use Measures for both Facts and Metrics. 
  • The Analytical Designer also uses the Calculated Measures term which is equivalent to Metrics

See also a brief summary of the key terms in our Knowledge Base article about GoodData Essential Concepts and Terms.

On a loosely related note, if you are looking for best practices for naming your measures then you may find the following article useful too: Best Practices: Naming Conventions for MAQL Measures.