What is Clustering in Tableau?

Ever looked at a scatter plot brimming with thousands of data points and wished you could automatically group them together to spot the hidden patterns? That's precisely where Tableau's clustering feature comes in. This article will show you what clustering is, why it's a powerful tool for analysis, and exactly how you can use it to find meaningful segments in your own data, step by step.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

What is Clustering? Your Data's Personal Organizer

In simple terms, clustering is an analysis technique that groups your data points into "clusters" where points in the same group are more similar to each other than to those in other groups. Think of it like a librarian automatically sorting a massive pile of returned books into distinct genres - fiction, history, science - without being told what the genres are ahead of time. The algorithm just looks at the characteristics of each book (its themes, writing style, etc.) and groups the similar ones together.

This is a form of unsupervised machine learning, which sounds more complex than it is. "Unsupervised" simply means you aren’t telling the model what the final groups should be. You just provide the data and the criteria for similarity, and it discovers the natural groupings for you.

Tableau uses a common clustering algorithm called k-means. While you don’t need to be a data scientist to use it, understanding the gist is helpful. Imagine your scatter plot is a map of customer locations. The k-means algorithm works by:

• Picking a few random central points on the map (the "k" in k-means).
• Assigning every single customer to their closest central point.
• Moving the central point to the true center of all the customers assigned to it.
• Repeating steps 2 and 3 until the cluster groupings are stable and no longer change.

The end result is a set of optimized clusters where each data point belongs to the group with the nearest center, revealing structures in your data that would be incredibly difficult, if not impossible, to find manually.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Why Use Clustering in Your Tableau Dashboards?

Clustering isn’t just a neat statistical trick, it provides tangible business value by turning raw data into organized, actionable segments. It helps you move from asking "what happened?" to understanding "who did it happen to?" and "where should we focus our efforts?"

Uncover Natural Customer Segments

This is one of the most popular uses for clustering. Instead of relying on broad demographic categories (like age or location), you can group customers based on their actual behavior. For instance, by clustering customers based on their sales volume and order frequency, you might uncover groups like:

• High-Value VIPs: Customers who buy expensive items frequently. They are perfect candidates for loyalty programs and exclusive offers.
• Deal Seekers: Customers who buy often but only with large discounts. This group might respond well to promotional campaigns but isn't contributing much to your profit margin.
• New & Promising: Customers with recent, low-value purchases. They could be targeted with welcome campaigns or product recommendations to encourage repeat buys.

Identify Market Opportunities and Underperformance

Clustering can be applied to geographical data to reveal regional patterns. If you cluster your sales territories based on metrics like profit ratio, sales growth, and market penetration, you might find:

• A cluster of high-performing territories that can serve as a model for the rest of the company.
• A group of underserved or struggling regions that require more resources, different strategies, or further investigation.
• An emerging market cluster that shows high growth potential, signaling an opportunity for expansion.

Optimize Product and Inventory Strategy

Do you know which of your products are typically bought together? Or which items appeal to the same customer base? By clustering products based on sales volume, seasonality, and profit margin, you can inform several key business decisions. For example, you could identify:

• Core Products: A cluster of consistently high-selling, high-margin items that are critical to your bottom line.
• Niche Seasonal Items: A group of products that sell really well, but only during specific times of the year. This helps with inventory planning and timely marketing.
• Loss Leaders: Products that might not be profitable on their own but are effective at driving traffic and sales of other items.

How to Create Clusters in Tableau: A Step-by-Step Guide

Creating clusters in Tableau is surprisingly straightforward. Let's walk through an example using the standard "Sample - Superstore" dataset that comes with Tableau. Our goal is to segment customers based on their total sales and profit.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Step 1: Set Up Your Visualization

Clustering needs a view that shows the individual data points you want to group. A scatter plot is the perfect starting point.

• Connect to the Sample - Superstore data source.
• Drag the Sales measure to the Columns shelf.
• Drag the Profit measure to the Rows shelf.
• To make sure we're looking at each individual customer, drag Customer Name from the Dimensions pane onto the Detail card in the Marks pane.

You should now have a scatter plot where each point represents a single customer, plotted according to their total sales and profit.

Step 2: Add the Cluster Model

With your scatter plot ready, you just need to tell Tableau to perform the clustering analysis.

• Navigate to the Analytics pane (it's located to the right of the Data pane at the top left).
• Under the "Model" section, you’ll see an option for Cluster. Simply drag this and drop it onto your chart area. A drop-box that says "Add a Cluster" will appear over the chart.

Step 3: Configure Your Clusters

The moment you drop the Cluster model onto the view, Tableau runs the analysis and a "Clusters" dialog box appears. Tableau will automatically group the data points for you, but you can fine-tune the settings here.

In the dialog box, you'll see two main components:

• Variables: These are the measures Tableau is using for the clustering algorithm. Since our view uses SUM(Sales) and SUM(Profit), those are added by default. You can drag additional fields into this box if you want to cluster based on more dimensions (e.g., you could add Discount to find clusters based on sales, profit, AND discount sensitivity).
• Number of Clusters: This is where you set 'k' - the number of groups you want Tableau to create. You can type in a number yourself, or leave it blank and let Tableau determine the optimal number of clusters for you. For your first attempt, it's often best to let Tableau decide.

For now, let’s leave the defaults as they are and simply close the dialog box. You'll immediately see your scatter plot transform, with each customer point now color-coded based on its assigned cluster.

Step 4: Analyze and Utilize Your New Cluster Field

Tableau doesn't just color your dots - it creates a brand new group field that you can use just like any other dimension in your workbook.

• In the Marks pane, you'll see a new field called Clusters on the Color card.
• In the Data pane on the left, you'll also see that same Clusters field listed. You can right-click this and rename the clusters to something more intuitive (e.g., 'High-Value Customers', 'Low Profitability/High Maintenance').

This new field is now a fully functional dimension. You can drag it into other visualizations to analyze these segments further. For instance, you could create a bar chart showing the average discount per cluster to see if unprofitable customers are the ones receiving the biggest discounts.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Interpreting Your Clusters: Are They Meaningful?

Tableau does the heavy lifting, but it's still your job as an analyst to determine if the clusters it found are statistically significant and make sense in a business context. Luckily, Tableau provides a handy tool for this.

Right-click the Clusters field on the Color card and select Describe Clusters.... This will open a new window with two tabs providing statistical details about the model it created.

The key part is the Models tab. This tab shows an F-statistic and a p-value for each variable used. Without getting too deep into the statistics, here’s what you should look for:

• High F-statistic: This number indicates how much separation there is between your clusters. A larger number is better, as it means the clusters are more distinct from one another.
• Low p-value (e.g., less than 0.05): The p-value tells you the probability that the differences between your clusters occurred by random chance. A very small p-value (close to 0) means it's extremely unlikely the results are random, giving you confidence that the clusters are statistically significant.

In our Superstore example, the p-value is effectively zero, which is a great sign! It confirms that our customer segments based on Sales and Profit are truly distinct and meaningful representations of different customer behaviors.

Final Thoughts

Tableau's clustering feature is an incredibly powerful yet simple way to uncover natural, data-driven segments hidden within your datasets. It empowers you to move beyond basic reporting to conduct true exploratory analysis, helping you discover customer archetypes, refine marketing strategies, and pinpoint operational strengths and weaknesses you never knew you had.

Building these analyses in tools like Tableau is an excellent skill, but the process of getting your data connected, cleaned, and visualized can still be a significant time sink. Creating custom dashboards often requires a steep learning curve. At Graphed, we automate the hard parts. We designed our platform so you can simply connect your data sources - like Shopify, Google Analytics, or Salesforce - and then ask for what you need in plain English. Instead of building a scatter plot manually, you could ask, "Build a dashboard segmenting my customers by sales and profit," and get a live, automated report in seconds, freeing you up to focus on strategy instead of setup.