How to Create a Dimension in Tableau

Creating custom dimensions is your secret weapon for unlocking deeper insights in Tableau. While the default fields in your dataset are a great starting point, the real magic happens when you start shaping, segmenting, and categorizing your data to answer specific business questions. This article will show you several practical and straightforward methods to create new dimensions from your existing data.

First, A Quick Refresher: Dimensions vs. Measures

Before creating new dimensions, it’s helpful to quickly recap the difference between dimensions and measures - the two fundamental types of data in Tableau.

Dimensions are qualitative, categorical data. Think of them as the “who, what, where, and when” of your dataset. They provide context and are the fields you use to slice and dice your information. Common examples include:

• Customer Name
• Product Category
• Region
• Order Date
• Campaign Name

In Tableau, dimensions are typically shown as blue pills in the Data Pane and indicate discrete (individually separate) values.

Measures, on the other hand, are quantitative, numerical data. They are the numbers you can perform mathematical operations on, like summing, averaging, or counting. Think of them as answering “how much?” questions. Examples include:

• Sales
• Profit
• Website Sessions
• Number of Orders
• Ad Spend

Tableau represents measures as green pills, which indicate continuous (forming an unbroken whole) values.

Understanding this distinction is crucial because it dictates how your visualizations are built. Using a measure where a dimension is needed (or vice versa) can lead to confusing charts and incorrect conclusions.

Why Would You Need to Create a New Dimension?

Your raw data isn't always perfectly formatted for analysis. Creating new dimensions helps you bridge that gap. You might want to:

• Categorize Records: Group similar items together, like combining dozens of individual ad campaign names into broader strategic buckets like "Awareness" or "Retargeting."
• Enrich Your Dataset: Add context that doesn't exist in a single field, such as creating a "Sales Region" dimension from a "State" field.
• Segment Data: Group a continuous measure into countable ranges, like creating age brackets ("18-24", "25-34") from individual customer ages.
• Clean Up Messy Data: Standardize inconsistent entries, like consolidating "fb," "Facebook Ads," and "facebook.com" into a single "Facebook" source.

Let's look at the most common methods for accomplishing these tasks.

Method 1: Creating Dimensions with Groups

Grouping is one of the easiest ways to create a new dimension. It allows you to manually combine different members of an existing dimension into a single category. Imagine you have a list of all your marketing campaign names, and you want to analyze performance by strategic goal rather than by individual campaign.

Here’s how you can group them:

1. In the Data pane on the left, find the dimension you want to group (e.g., Campaign Name) and right-click on it.
2. In the context menu, select Create > Group...
3. This opens a dialog box listing all the members of your dimension. Select the campaign names you want to categorize together (you can hold Ctrl or Cmd to select multiple).
4. Once they're highlighted, click the Group button. Tableau will create a new group with the selected members. You can then click the group's name to rename it something meaningful, like "Brand Awareness Campaigns."
5. Repeat this process for your other campaigns, creating groups like "Lead Generation" and "Seasonal Promotions."
6. You can also check a box labeled "Include 'Other'". This useful option lumps any ungrouped members into a catch-all category, ensuring nothing gets missed.

When you click OK, a new dimension (Campaign Name (group)) appears in your Data pane. You can now drag this field into your view to see your measures neatly organized by your new custom categories.

Method 2: Using Bins to Segment Continuous Data

What if you want to group a measure, not a dimension? Bins are the answer. Bins allow you to take a continuous measure like Age, Order Amount, or Website Session Duration and convert it into a set of discrete ranges, or "bins." This is perfect for creating histograms to see the distribution of your data.

For example, let's categorize customers by age to understand your key demographics.

1. Find your continuous measure in the Data pane (e.g., Customer Age).
2. Right-click the measure and choose Create > Bins...
3. A new window will appear. Tableau will suggest a default bin size based on your data's range. For example, if your ages range from 18 to 75, Tableau might suggest a bin size of 10.
4. You can edit this number. A size of 10 would create groups like "10-19," "20-29," etc. You can adjust this to 5 to get more granular groups if needed.
5. Once you're happy with the size, click OK.

A new dimension called Customer Age (bin) will show up in your Data Pane. This new blue pill allows you to analyze the count of customers, average order value, or any other measure by these newly created age brackets.

Method 3: Unleashing the Power of Calculated Fields

Calculated fields are the most flexible and powerful way to create new dimensions. They allow you to apply custom logic and formulas to your existing data, much like using formulas in Excel or Google Sheets. This opens up endless possibilities for segmentation, cleaning, and enrichment.

Creating Categories with IF/ELSE or CASE Statements

Conditional logic is perfect for creating new categories based on certain criteria. Imagine you want to segment individual orders into "Small Order" or "Large Order" based on their value.

Here's how to do it with an IF/THEN statement:

1. In the Data pane, right-click anywhere in the blank space and select Create Calculated Field...
2. Give your new dimension a name, like Order Size.
3. In the formula editor, type the following logic:

IF [Order Value] > 200 THEN "Large Order"
ELSE "Small Order"
END

1. At the bottom of the editor, a message will say "The calculation is valid." Click OK.

You’ve just created a new dimension that labels every row in your data based on its order value. This allows you to quickly see how many large versus small orders you receive or compare their profitability.

For more complex categorization with several conditions, a CASE statement often works better and is easier to read. For example, creating a "Region" dimension based on "State" data:

CASE [State]
WHEN "WA" THEN "West"
WHEN "OR" THEN "West"
WHEN "CA" THEN "West"
WHEN "NY" THEN "East"
WHEN "MA" THEN "East"
ELSE "Central"
END

Segmenting by Date Properties

Calculated fields are also fantastic for working with dates. Let's say you want to create a cohort identifier for each customer based on the year they made their first purchase.

• Field Name: Customer Acquisition Year
• Formula:

STR(YEAR([First Purchase Date]))

This simple formula extracts the year from a date field and converts it into a string (text), creating a dimension you can use to compare customer behavior year over year.

Method 4: Quick Fixes for Common Data Issues

Sometimes, all you need is a quick adjustment rather than a complex calculation. Tableau has built-in tools for two very common scenarios.

Converting a Measure to a Dimension

Tableau sometimes misinterprets a field. Numeric fields like Zip Code, Customer ID, or even Order ID are often imported as measures because they contain numbers. But you would never want to calculate the SUM of zip codes! These are really identifiers - categories deserving to be dimensions.

The fix is incredibly easy:

1. Find the field in the Measures section of the Data pane (it will be green).
2. Simply drag and drop it into the Dimensions section above.

That's it. Tableau will convert it into a discrete blue pill, and you can now use it correctly for bucketing and categorizing.

Splitting a Column for More Granularity

Often, a single field contains multiple pieces of useful information separated by a delimiter like a comma, hyphen, or underscore. A common example is a marketing campaign ID like Facebook_Brand-Awareness_Q4-2023.

Analyzing this as a single field is hard. It would be much more useful to split it into three separate dimensions: Source, Objective, and Timeframe. Here’s how to do that with Tableau's split function:

1. Right-click the dimension in the Data pane (e.g., Campaign ID).
2. Go to Transform > Split.

Tableau will automatically try to detect the delimiter and create new fields like Campaign ID - Split 1, Campaign ID - Split 2, and so on. If your naming convention is less consistent, you can choose Transform > Custom Split... to manually specify the delimiter and how many splits to create.

Final Thoughts

Creating custom dimensions is an essential skill for turning raw data into meaningful business stories in Tableau. Whether you're grouping campaigns, bucketing customers with bins, or building custom logic with calculated fields, you can now structure your data precisely for the insights you need.

If you're tired of manually digging through reports and find tools like Tableau require more setup than you have time for, you're not alone. We created Graphed to automate the tedious work of wrangling data. Just connect your marketing and sales sources in a few clicks, then ask for the charts and dashboards you want using plain English. Our AI-powered analyst builds everything in seconds, so you can stop wrestling with tools and start making better decisions, faster.