LOD vs table calculations in Tableau - Compared

Introduction

This feature helps you understand the difference between LOD expressions and table calculations in Tableau. It solves the problem of choosing the right method to calculate values at different levels of detail or order in your data visualization.

When you want to calculate a value ignoring the view's current filters or dimensions, like total sales per customer regardless of the current chart grouping.

When you need to compute running totals or moving averages that depend on the order of rows in your table.

When you want to fix a calculation at a specific level, such as average sales per region even if the view shows data by city.

When you want to calculate percent of total based on the displayed data in the current view.

When you need to compare values across different levels of detail without duplicating data.

Steps

Step 1: Open

- Tableau Desktop and load your data source

Your data appears in the Data pane on the left side

Step 2: Create a new calculated field

- Data pane → right-click → Create Calculated Field

A calculation editor window opens

Step 3: Type an LOD expression like { FIXED [Customer]: SUM([Sales]) }

- Calculation editor

The calculation fixes sales sum per customer regardless of view filters

💡 Use curly braces and keywords FIXED, INCLUDE, or EXCLUDE for LOD expressions

Step 4: Drag the new LOD calculated field to Rows or Columns shelf

- Worksheet view

The chart updates showing sales fixed at customer level

Step 5: Create another calculated field for table calculation like RUNNING_SUM(SUM([Sales]))

- Data pane → right-click → Create Calculated Field

A calculation that sums sales cumulatively in the table

Step 6: Drag the table calculation field to the view

- Worksheet view

The chart updates showing running total of sales based on table order

Step 7: Right-click the table calculation field → Edit Table Calculation to adjust addressing and partitioning

- Marks card or shelf

You can control how the calculation moves across rows or columns

Before vs After

Before

View shows total sales by region and city with no special calculations

After

View shows sales fixed per customer using LOD and running total of sales using table calculation

Settings Reference

LOD Expression Types

📍 Calculation editor when writing calculated fields

Controls how the level of detail is fixed or adjusted relative to the view

Default: No default, user defines

Table Calculation Compute Using

📍 Edit Table Calculation dialog

Defines the direction and scope of the table calculation

Default: Table (Across)

Restarting Every

📍 Edit Table Calculation dialog

Sets where the table calculation resets its calculation

Default: None

Common Mistakes

Using LOD expressions when a running total or order-based calculation is needed

LOD expressions ignore table order and cannot compute running totals

Use table calculations like RUNNING_SUM for order-dependent calculations

Using table calculations to fix values at a different level of detail

Table calculations depend on the current view and cannot fix aggregation levels independently

Use LOD expressions with FIXED to calculate values independent of the view

Not adjusting the Compute Using setting for table calculations

Table calculations may produce incorrect results if addressing and partitioning are wrong

Edit Table Calculation and set Compute Using to the correct dimensions

Summary

LOD expressions fix calculations at specific data levels regardless of view filters or dimensions.

Table calculations compute values based on the order and layout of data in the current view.

Choose LOD for level of detail control and table calculations for order-dependent computations.

Practice

1. What is the main difference between LOD expressions and table calculations in Tableau?

easy

A. LOD expressions fix calculations at a specific data level, ignoring some filters, while table calculations work on the data shown in the current view.

B. LOD expressions only work on aggregated data, while table calculations only work on raw data.

C. LOD expressions are faster than table calculations in all cases.

D. Table calculations ignore filters, but LOD expressions depend on the table layout.

Solution

1. Step 1: Understand LOD expressions

   LOD expressions fix the calculation at a certain level of detail, ignoring some filters applied to the view.

2. Step 2: Understand table calculations

   Table calculations work on the data currently displayed in the view and depend on how the table is laid out.

3. Final Answer:

   LOD expressions fix calculations at a specific data level, ignoring some filters, while table calculations work on the data shown in the current view. -> Option A

4. Quick Check:

   LOD fixes level, table calculations depend on view [OK]

Hint: LOD fixes level, table calculations depend on view layout [OK]

Common Mistakes:

• Thinking table calculations ignore filters
• Believing LOD always works on raw data
• Assuming LOD is always faster

2. Which of the following is the correct syntax for a fixed LOD expression in Tableau to calculate total sales per region?

easy

A. { EXCLUDE [Region] : SUM([Sales]) }

B. { INCLUDE [Region] : SUM([Sales]) }

C. { FIXED [Region] : SUM([Sales]) }

D. SUM([Sales]) FIXED BY [Region]

Solution

1. Step 1: Recall fixed LOD syntax

   The fixed LOD syntax is { FIXED [Dimension] : Aggregation }.

2. Step 2: Match syntax to question

   { FIXED [Region] : SUM([Sales]) } matches the correct syntax for total sales per region.

3. Final Answer:

   { FIXED [Region] : SUM([Sales]) } -> Option C

4. Quick Check:

   Fixed LOD uses FIXED keyword and curly braces [OK]

Hint: Fixed LOD always starts with { FIXED ... } [OK]

Common Mistakes:

• Using INCLUDE or EXCLUDE instead of FIXED
• Missing curly braces
• Incorrect keyword order

3. Given a view showing sales by category and sub-category, what will the following LOD expression return?
{ FIXED [Category] : SUM([Sales]) }

medium

A. Total sales for each sub-category ignoring category

B. Total sales for each category repeated for all sub-categories within it

C. Total sales for the entire dataset ignoring category and sub-category

D. Running total of sales by sub-category

Solution

1. Step 1: Understand FIXED by Category

   The expression fixes sales at the category level, ignoring sub-category detail.

2. Step 2: Effect on sub-category rows

   Each sub-category row will show the total sales of its parent category, repeated.

3. Final Answer:

   Total sales for each category repeated for all sub-categories within it -> Option B

4. Quick Check:

   Fixed LOD repeats category total per sub-category [OK]

Hint: Fixed LOD repeats fixed level value for lower detail rows [OK]

Common Mistakes:

• Thinking it sums only sub-category sales
• Confusing FIXED with INCLUDE or EXCLUDE
• Assuming running total behavior

4. You created a table calculation for running total of sales but it shows incorrect results. Which of the following is the most likely cause?

medium

A. The filter is applied before the LOD calculation.

B. The LOD expression used is FIXED instead of INCLUDE.

C. The data source is missing a join condition.

D. The table calculation is not set to compute using the correct dimension.

Solution

1. Step 1: Understand table calculation behavior

   Table calculations depend on the table layout and the dimension used for computation.

2. Step 2: Identify common error

   If the running total is wrong, often the compute using dimension is set incorrectly.

3. Final Answer:

   The table calculation is not set to compute using the correct dimension. -> Option D

4. Quick Check:

   Table calcs need correct compute using dimension [OK]

Hint: Check 'compute using' dimension for table calculations [OK]

Common Mistakes:

• Confusing LOD and table calculation errors
• Blaming data joins for table calc issues
• Ignoring table layout impact

5. You want to show the percent of total sales by region in a view that filters by year. Which approach is best to ensure the percent always reflects all years, ignoring the year filter?

hard

A. Use a FIXED LOD expression to calculate total sales by region ignoring the year filter, then divide sales by this fixed total.

B. Use a table calculation for percent of total sales by region, which automatically ignores filters.

C. Apply the year filter after creating a table calculation for percent of total sales.

D. Use an INCLUDE LOD expression including year to calculate sales.

Solution

1. Step 1: Understand filter impact on calculations

   Table calculations and INCLUDE LOD respect filters, so year filter affects them.

2. Step 2: Use FIXED LOD to ignore year filter

   FIXED LOD can ignore filters like year, fixing total sales by region across all years.

3. Step 3: Calculate percent of total

   Divide sales by the fixed total sales to get percent ignoring year filter.

4. Final Answer:

   Use a FIXED LOD expression to calculate total sales by region ignoring the year filter, then divide sales by this fixed total. -> Option A

5. Quick Check:

   FIXED LOD ignores filters, perfect for fixed percent totals [OK]

Hint: Use FIXED LOD to ignore filters for fixed percent totals [OK]

Common Mistakes:

• Using table calculations that respect filters
• Using INCLUDE LOD which respects filters
• Applying filters after calculations incorrectly