Exploratory Analysis and Data Visualization
Choosing the right chart
This domain rewards matching a question to a chart type and reading Tableau's aggregation logic. Know when to use bar charts (compare categories), line charts (trend over time), scatter plots (relationship between two measures), histograms (distribution of one measure), heatmaps, and box plots. Understand aggregation (SUM, AVG, COUNTD, MIN/MAX) and how granularity is set by the dimensions on the view - the level of detail determines each mark.
Calculations and table calculations
Be comfortable with row-level vs. aggregate calculations, LOD expressions (FIXED, INCLUDE, EXCLUDE), and table calculations (running total, percent of total, rank, difference). Know that LODs compute at a specified grain independent of the view, while table calculations operate on what is already in the view and depend on addressing/partitioning. Understand quick table calculations and how filters interact with them.
Traps to watch
- Using a table calc when an LOD is needed (or vice versa) - a filter can change a table calc result but not a FIXED LOD.
- Comparing averages of averages instead of the true aggregate.
- Misreading a scatter plot whose marks are aggregated rather than row-level.
- Forgetting that a reference line or trend line does not change the underlying marks.
Worked example
To show each customer's total sales as a percent of all sales regardless of the category filter on the sheet, use { FIXED : SUM([Sales]) } in the denominator so the total ignores the view's category dimension. A table calc would shrink with the filter.