LOD Expressions in Tableau: How Does it Work? [Guide For Beginners]

Tableau is one of the most popular and extensively used data visualization tools in the world. The tool’s primary aim is to enable you to see and understand the connotations of data. The features of Tableau are designed to transform data insights into meaningful and actionable business insights. Learn more about tableau data visualization.

Since Tableau focuses on data calculations, the better you can express ideas in a calculation language, the better meaning you will be able to generate from the data at hand. This is precisely why Tableau Level of Detail (LOD) expressions was introduced. To learn more about Tableau and other data related tools, check out our data science online certifications from top universities.  

The Tableau Level of Detail (LOD) expressions was introduced in Tableau 9.0. These LOD expressions allow you to express powerful concepts by using simple statements.

What are Tableau LOD Expressions?

Tableau LOD expressions are powerful tools that you can use to find answers to questions that involve multiple levels of granularity in a single visualization. Using LOD expressions, you can compute aggregations that are not at the level of detail of visualization. You can incorporate the aggregation values within the data visualizations in arbitrary ways. Furthermore, LOD expressions give you more control to determine the level of granularity in your computations. 

Now, without further ado, let’s get into a detailed discussion on Tableau LOD expressions.

Understanding LOD Expression Syntax

In Tableau, LOD expressions adhere to a particular syntax that enables users to specify the level of detail for calculations. The LOD keyword precedes the aggregating function and the dimension(s) to be taken into account in the syntax. For example:

{FIXED [Dimension]: Aggregation}
{INCLUDE [Dimension]: Aggregation}
{EXCLUDE [Dimension]: Aggregation}

FIXED Level of Detail Expressions

Tableau’s set LOD expressions let us specify a set level of detail without taking into account the level of granularity of the visualisations. This indicates that regardless of the dimensions contained in the view, the calculation will remain constant. The “FIXED” keyword is used to indicate Fixed LOD in Tableau. For instance:

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

Regardless of any additional dimensions or filters used, this formula will calculate the sum of sales for each region.

INCLUDE Level of Detail Expressions

Tableau’s INCLUDE LOD expressions let us specify extra dimensions to include in the calculation. The keyword “INCLUDE” is used to identify these expressions. For example:

{INCLUDE [Region], [Product Category]: AVG([Profit])}\

Regardless of any additional dimensions or filters, this phrase will calculate the average profit for each location and product category.

EXCLUDE Level of Detail Expressions

Tableau EXCLUDE LOD expressions let users take particular dimensions out of the calculation. The “EXCLUDE” keyword designates certain expressions. Here’s an example:

{EXCLUDE [Product Sub-Category]: MAX([Sales])}

The maximum sales, omitting the product subcategory dimension, will be determined by this equation.

Tableau: Tables vs. LOD Expressions

LOD expressions offer a more potent and adaptable mechanism to do computations at many levels of detail than Tableau’s flexibility in generating tables to display data. Users can go around the constraints of conventional tables and gain deeper insights into their data by utilising LOD expressions.

Tableau Tables:

• Present information at the finest level possible.
• Based on dimensions and measurements, summarise the data.
• Utilise aggregations to provide simple calculations.

Expressions for LOD

• Regardless of the visualisation, allow calculations at particular degrees of granularity.
• Permit the addition or exclusion of dimensions to concentrate on particular analytical elements.
• Increased flexibility for sophisticated analysis and complex calculations.

Tableau LOD Expressions

The top five Tableau LOD expressions are:

1. FIXED LOD 

This expression can compute values using specified dimensions without reference to any other dimensions present in the view. This LOD expression does not consider any filter in the view apart from context filters, data source filters, and extract filters.

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

2. INCLUDE LOD 

This expression computes values using specified dimensions along with the dimensions that are in the view. This LOD expression is best suited for situations where you wish to include a dimension that is not in the view.

{ INCLUDE [Customer Name] : SUM([Sales]) }

3. EXCLUDE LOD

These expressions are designed to remove dimensions from an expression. Essentially, they subtract dimensions from the level of detail of the view. 

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

Read: Tableau Developer Salary in India

4. Row Level 

Usually, in Tableau, the expressions that reference the unaggregated data source columns are computed for each row in the underlying table. Thus, the dimensionality of the expression here is row level. Below is an example of a row level expression:

[Sales] / [Profit]

This expression will evaluate each row of the database, wherein the Sales value in each row will be divided by the Profit value in that particular row to produce a new column containing the profit ratio.

You can save the computation of this definition with the name [ProfitRatio]. When you drag the name from the data pane to a shelf, Tableau will automatically aggregate the calculated field for the view:

SUM[ProfitRatio]

5. View Level 

In this LOD, the expressions referencing the aggregated data source columns are computed at the dimensionality defined by the dimensions in the view. Thus, the dimensionality of the expression is view level. Here’s an example of a view-level expression is:

SUM(Sales) / SUM(Profit)

You can drag this computation from the data pane to a shelf or you can also type it directly on a shelf as an ad-hoc calculation, and Tableau will enclose it in an AGG function like so:

AGG(SUM(Sales) / SUM(Profit))

This is called as Aggregate Calculation (ACG).

Also read: Tableau Architecture

6. Table-Scoped

Apart from the row and view level of detail expressions, you can also define a Tableau LOD expression at the table level without using any scoping keywords. Here’s a simple example of Table-Scoped LOD. If you write the following expression {MIN([Order Date])}, it will return the minimum (earliest) order date for the entire table.

Table-Scoped LOD is the equivalent of a FIXED LOD having no dimension declaration. So, the expression will be as follows:

{FIXED : MIN([Order Date])}

How to create LOD expressions?

Here’s how you can create Tableau LOD expressions in three simple steps!.

Step 1: Configure the visualization

• First, you must open Tableau Desktop and connect to the Sample-Superstore saved data source.
• Now, you must navigate to a new worksheet.
• Go to the Data pane section under Dimensions and drag Region to the Columns shelf.
• Lastly, go to the Data pane under Measures and drag Sales to the Rows Shelf. 

After completing these four steps, you will see a bar chart showing the sum of sales for each region, like so:

Source 

Step 2: Create the LOD expression

Now that you’ve seen the sum of sales per region, you can create a LOD expression to see the average sales per customer for each region. To do so, you can use a LOD expression in the following way:

• First, select Analysis, and then go to Create Calculated Field.
• In the Calculation editor that appears before you, input the following:

  – Name the calculation “Sales Per Customer.”

– Enter the following LOD expression: { INCLUDE [Customer Name] : SUM([Sales]) }

• When done, click on OK. Now, the newly created LOD expression will be added to the Data pane under Measures.

Step 3: Use the newly created LOD expression in the visualization

• Go to the Data pane under Measures and drag Sales Per Customer to the Rows shelf and place it to the left of the SUM(Sales).
• Now go to the Rows shelf and right-click Sales Per Customer and select Measure (Sum) > Average.
• The final result will be the sum of all sales and the average sales per customer for each region.

You can now see both the sum of all sales and the average sales per customer for each region. For instance, in the bar chart, it is visible that in the Central region, the sales totaled approximately US$ 500,000, with an average sale for each customer being approximately US$ 800.

Source

upGrad’s Exclusive Data Science Webinar for you –

Transformation & Opportunities in Analytics & Insights

Tableau LOD: Limitations 

All Tableau LOD expressions have the following limitations and constraints:

• The LOD expressions that reference floating-point measures can depict unpredictable behavior when used in a view that demands a comparison of the values in the expression.
• For LOD expressions, when you reference a parameter in a dimensionality declaration, you must always use the parameter name and not the parameter value.
• When you perform data blending, you must ensure that the linking field from the primary data source is in the view before you can use a level of detail expression from the secondary data source.

Conclusion

If you found this interesting, feel free to experiment with Tableau LOD expressions to know their unique quirks and advantages!

Join the Executive PG Programme in Data Science course at upGrad today and be part of the growing developers. Get in touch with us for more information and one of our executives will take you through the process.