How Do You Avoid Aggregation in Tableau?

One of the first things you'll notice in Tableau is that it loves to summarize your data. When you drag a "Sales" field into your view, it instantly becomes "SUM(Sales)." This automatic aggregation is a powerful feature for analyzing large datasets, but it can be confusing when you want to see the raw, row-level details. This article will walk you through how to turn off or control aggregation in Tableau to see your data just the way you want it.

Graphed

Your AI Data Analyst to Create Live Dashboards

Connect your data sources and let AI build beautiful, real-time dashboards for you in seconds.

Build Your First Dashboard

Why Does Tableau Aggregate Data by Default?

Tableau is built for visual analytics, not as a replacement for spreadsheets. Its primary job is to help you spot trends, patterns, and outliers in large volumes of data. To do this efficiently, it automatically groups and summarizes your numerical data (measures) based on the categories (dimensions) you're examining.

Imagine you have a dataset with 100,000 individual sales transactions. If you wanted to compare sales across three product categories, you wouldn't want Tableau to draw 100,000 marks on your screen. You'd want to see three summary bars - one showing the total sales for each category. This default aggregation (usually with SUM, but sometimes AVG, MIN, or MAX) makes that possible and fast.

Functions you commonly see are:

• SUM(): Adds all the values together.
• AVG(): Calculates the average of the values.
• MIN(): Finds the smallest value.
• MAX(): Finds the largest value.
• COUNT(): Counts the number of non-null records.
• COUNTD(): Counts the number of distinct (unique) records.

While this is incredibly useful for high-level analysis, there are times you need to drill down into the individual records. Let's cover the methods to do just that.

AI for Data Analysis

Crash Course

Graphed

Free PDF Guide

AI for Data Analysis Crash Course

Learn how to get AI to do data analysis for you — the best tools, prompts, and workflows to go from raw data to insights without writing a single line of code.

Download Free Guide

The Foundation: Understanding Dimensions and Measures

Before changing any settings, it's vital to know the difference between Dimensions and Measures. This is the core logic behind Tableau's aggregation behavior.

• Dimensions: These are qualitative, categorical fields that you can use to slice and dice your data. Think of them as the buckets for your numbers. Examples include Customer Name, Order Date, Product Category, and Region. In the data pane, they are usually colored blue.
• Measures: These are quantitative, numerical fields that can be measured and aggregated. They are the numbers that go into the buckets. Examples include Sales, Profit, Quantity, and Discount. They are typically colored green in the data pane.

Tableau aggregates Measures based on the Dimensions present in your view. If you have SUM(Sales) on the Rows shelf and Region on the Columns shelf, Tableau calculates the sum of sales for each region separately. The key to controlling aggregation is managing which dimensions define your level of detail.

Method 1: The Direct Approach - Disaggregating Measures

The most straightforward way to see every single data point is to tell Tableau to stop aggregating measures altogether. This is perfect for situations like creating scatter plots where you need one mark for every row in your data.

Let's say you want to build a scatter plot to see the relationship between Sales and Profit for every single order. If you drag Profit to the Rows shelf and Sales to the Columns shelf, you'll initially get a single mark representing the SUM of all profits and the SUM of all sales.

To fix this, follow these simple steps:

1. Navigate to the Analysis menu at the top of the screen.
2. Uncheck the option that says Aggregate Measures.

Instantly, the view will change. Instead of one aggregated mark, you'll see a cloud of marks, with each mark representing one row from your data source. This view displays disaggregated data with one data point (or mark, in Tableau terms) for each record.

Heads Up: While simple, this method can have major performance implications. If your dataset has millions of rows, asking Tableau to draw a mark for each one can dramatically slow down your dashboard or even cause it to freeze. Use this feature wisely, often in combination with filters to limit the data displayed.

Graphed

Your AI Data Analyst to Create Live Dashboards

Connect your data sources and let AI build beautiful, real-time dashboards for you in seconds.

Build Your First Dashboard

Method 2: The Pro's Choice - Adding More Detail with Dimensions

Instead of turning off aggregation entirely, a more flexible and common approach is to increase the level of detail in your view. This lets you see more granular data points while still taking advantage of Tableau's performance optimizations.

You control the level of detail by adding more dimensions to your view, specifically to the Detail shelf on the Marks card.

Let’s return to our example of SUM(Sales) by Category. This gives you three bars. Now, what if you want to see the individual orders that make up the total sales for the Technology category? You can accomplish this by dragging a more granular dimension, such as Order ID, onto the Detail shelf.

The moment you do this, the "Technology" bar will fracture into thousands of tiny marks, where each mark represents a unique Order ID. The aggregation is still SUM(Sales), but now it’s calculated for each Order ID within that category. You’re effectively seeing the raw sales number for each order because the Order ID brings the view down to the individual order-level data.

This approach offers several benefits:

• It maintains aggregation, which is better for performance.
• It provides flexibility to decide exactly how granular your analysis should be.
• It helps you create more insightful, layered visualizations without abandoning summaries.

Method 3: Advanced Control with LOD Expressions

For ultimate control over aggregation, advanced users can turn to Level of Detail (LOD) expressions. These calculations let you compute an aggregation at a level of granularity that's independent of the dimensions in your view.

Let's consider a simple use case where understanding the difference between LOD expressions and our other solutions can be powerful.

Imagine that you want to show how a single Sub-Category performs in comparison to its overarching Parent Category. Here’s how this process would work:

First, create a calculated field, which we'll call Category Sales in this example. Use a FIXED LOD expression, as follows:

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

This LOD expression tells Tableau to compute SUM(Sales) for each unique Category. You can then use the Category Sales measure in your visualizations, knowing that its level of aggregation will not be affected by other filters.

This method provides a straightforward way to identify how much of a Category's Sales can be attributed to an individual Sub-Category.

For example, build a view with Sub-Category on the Rows shelf and both SUM(Sales) and Category Sales on the Columns shelf to compare them directly.

AI for Data Analysis

Crash Course

Graphed

Free PDF Guide

AI for Data Analysis Crash Course

Learn how to get AI to do data analysis for you — the best tools, prompts, and workflows to go from raw data to insights without writing a single line of code.

Download Free Guide

Final Thoughts

Controlling data aggregation is a fundamental skill for becoming proficient in Tableau. Whether you’re turning it off completely via the Analysis menu or increasing granularity by adding dimensions to the Marks card, these techniques enable you to see the data at the desired level of detail.

At Graphed, our perspective is that you shouldn’t have to learn complex tools to extract insights from your data. We simplify data analysis processes so you can focus on interpreting results. Our platform provides visualizations in a single-page view, saving you time and effort. Best of all, Graphed offers a free solution for users with fewer than three data connections.

We aim to make your time count, enabling you to explore new possibilities with no learning curve. Start using our tool today and discover the potential of easy, intuitive data visualization.

At Graphed, our goal is to create a product that gives you back time, so you can spend it helping others or focusing on your own growth. Experience the convenience of our platform, whether you choose the free version with limited connections or our custom pricing option. Your data will be instantly accessible whenever your current project requires it.