Salesforce · Analytics-101
Key Concepts Explained
Dimensions and measures represent the fundamental categorization of data fields in Tableau, and understanding the distinction is essential for creating meaningful visualizations.
Dimensions and Measures
Dimensions and measures represent the fundamental categorization of data fields in Tableau, and understanding the distinction is essential for creating meaningful visualizations.
Dimensions are categorical fields containing qualitative values. They describe characteristics or attributes — customer names, product categories, geographic locations, date parts like year and month. Dimensions segment data into groups, enabling comparison and drill-down. When dragged to shelves, dimensions typically create headers and labels that organize the view.
Measures are quantitative fields containing numerical values that can be aggregated. They represent things you can count, sum, or average — sales revenue, quantity sold, profit, customer age. Measures are the numbers you want to analyze, appearing in green by default as continuous fields.
Use dimensions to create structure and categories in your visualization. Use measures to provide the quantitative values that fill those categories. Most visualizations require both: dimensions determine the labels and groupings, while measures determine the heights of bars, positions of points, or intensities of colors.
When you place a measure on a shelf, Tableau aggregates it using SUM, AVG, MIN, MAX, COUNT, etc. This aggregation level depends on the dimensions present. If you have sales by region, sales sums for each region. Add product category, and sales sums for each region-category combination.
Benefits of proper classification include accurate aggregations, appropriate default visualization types, and intuitive analytical workflows. Tableau's interface is designed around this distinction.
A numeric postal code might default to a measure, but treating it as a dimension makes more sense since postal codes are categories, not quantities to sum. Users must understand when to override default assignments. This is a classic exam scenario — they give you a field that's technically numeric but conceptually categorical.