Tableau Level of Detail Expressions

Level of Detail (LOD) expressions let you control the granularity of a calculation independently from the chart's current view. A standard Measure aggregates at whatever level your Dimensions define. An LOD expression locks the aggregation to a specific level you choose — regardless of what is on screen.

The Problem LODs Solve

Imagine you want to show each customer's order on a chart, but you also want to display the customer's total annual sales next to each order. The chart is at the order level. The total annual sales is at the customer + year level. Without LOD, Tableau cannot hold both levels at once. LOD expressions solve this.

Analogy: School Grades

You are viewing data at the individual test level:
  Student | Subject | Test Score
  Alice   | Math    | 85
  Alice   | Science | 78
  Alice   | English | 90

You want to add: Alice's OVERALL average across all subjects = 84.3
even while the chart is still showing individual subjects.

LOD Expression:
{FIXED [Student] : AVG([Test Score])}
→ Always computes AVG per student, regardless of chart level
→ Alice gets 84.3 on every row, even on the individual test rows

The Three Types of LOD Expressions

FIXED

FIXED computes at a level you define. It ignores Dimension filters (but not context filters). Use FIXED when you need a calculation to always group at a specific field, no matter what the chart shows.

Syntax:
{FIXED [Dimension] : Aggregation([Measure])}

Example — Total Sales per Customer (always at customer level):
{FIXED [Customer Name] : SUM([Sales])}

Usage:
Drag this field to any chart. Even if your chart groups by Order ID,
this field always returns the customer's total, not the order's total.

INCLUDE

INCLUDE adds a Dimension to the calculation that is not in the view. It computes at a more detailed level than the current chart. Use INCLUDE when you want to compute at a finer grain and then aggregate that result up to the chart's level.

Syntax:
{INCLUDE [Dimension] : Aggregation([Measure])}

Example — Average of customer totals per region:
{INCLUDE [Customer Name] : SUM([Sales])}

Usage:
If your chart is at the Region level, this computes SUM(Sales)
per Customer within each region, then returns the AVG of those
customer sums — giving average customer value per region.

Without INCLUDE: SUM(Sales) / COUNT(Customers) — divides totals
With INCLUDE: properly averages each customer's individual total

EXCLUDE

EXCLUDE removes a Dimension from the calculation even if it appears in the view. Use EXCLUDE when you want a calculation to ignore a specific Dimension and aggregate at a broader level.

Syntax:
{EXCLUDE [Dimension] : Aggregation([Measure])}

Example — Total Sales for the entire Region, ignoring Category:
{EXCLUDE [Category] : SUM([Sales])}

Usage:
Your chart has both Region and Category in Rows.
Normally SUM(Sales) gives sales per Region+Category combination.
With EXCLUDE [Category], it gives the Region total on every row —
useful for computing each category's percentage of its region total.

Diagram: Comparing FIXED, INCLUDE, EXCLUDE

Chart Dimensions: Region, Category

                 | SUM(Sales)    | FIXED [Region] | EXCLUDE [Category]
-----------------+---------------+----------------+-------------------
East Furniture   | $500          | $1,800         | $1,800
East Technology  | $800          | $1,800         | $1,800
East Office Sup  | $500          | $1,800         | $1,800
West Furniture   | $300          | $900           | $900
West Technology  | $600          | $900           | $900

SUM(Sales)        = changes by Region + Category
FIXED [Region]    = locked to Region total (ignores Category)
EXCLUDE [Category]= same as FIXED Region here — removes Category

Practical Use Case: Percent of Region Total

A common dashboard need: show each category's percentage of its region's total sales. This requires two different levels simultaneously.

Step 1 — Create FIXED expression for Region Total:
Region Total Sales = {FIXED [Region] : SUM([Sales])}

Step 2 — Create Percent of Region field:
Percent of Region = SUM([Sales]) / SUM([Region Total Sales])

Step 3 — Format as percentage and drag to Labels

Result:
  East Furniture   — $500 — 28% of East total
  East Technology  — $800 — 44% of East total
  East Office Sup  — $500 — 28% of East total

LOD Expressions and Filters

FIXED expressions ignore Dimension filters. This is intentional — FIXED always computes at its defined level regardless of what the view shows. To make a FIXED expression respect a filter, promote that filter to a Context Filter. Right-click the filter on the Filters shelf and select "Add to Context." Context filters run before LOD expressions.

When to Use Each Type

Summary

LOD expressions control which level Tableau uses for a calculation, independently of the chart's current Dimensions. FIXED locks the level to specific fields. INCLUDE adds a finer Dimension to the calculation. EXCLUDE removes a Dimension from the calculation. LODs are powerful for computing percent of total at a parent level, comparing individual values to group averages, and mixing different granularities in one view — all without changing the chart's structure.