How to Make a Scatter Plot in Tableau with AI

A scatter plot is one of the most effective ways to spot a relationship between two different numbers - like how your ad spend influences sales, or whether bigger discounts lead to higher order quantities. Building one in Tableau is a fundamental skill, but there’s a much faster way to get the job done. This guide will walk you through the traditional steps to create a scatter plot in Tableau and then show you how AI is changing the game for data analysis.

What Is a Scatter Plot?

A scatter plot, or scatter chart, displays values for two numerical variables as a collection of dots. The position of each dot on the horizontal (X-axis) and vertical (Y-axis) axes indicates its value for each of the two variables. It's the perfect chart for answering questions about relationships and correlations.

For example, you could use a scatter plot to find answers to questions like:

Does spending more on marketing campaigns lead to a proportional increase in website conversions?
Is there a relationship between the price of a product and the number of units sold?
Do sales reps who make more calls a day also close more deals?
Are higher customer satisfaction scores correlated with higher customer lifetime value?

By plotting these data points, you can visually identify patterns. If the dots trend upwards from left to right, you have a positive correlation (as one variable increases, the other tends to increase). If they trend downwards, you have a negative correlation (as one increases, the other decreases). If the dots are scattered randomly with no clear trend, there’s likely no correlation.

Creating a Scatter Plot in Tableau: The Traditional Approach

Let's build a classic scatter plot in Tableau to analyze profitability. We’ll use the "Sample - Superstore" dataset that comes with Tableau, which makes it easy to follow along. Our goal is to see the relationship between Sales and Profit for each customer.

Step 1: Connect to Your Data

First, open Tableau Desktop and under "Saved Data Sources," select Sample - Superstore. Tableau will load the data and take you to the worksheet view.

Step 2: Add Measures to Columns and Rows

A scatter plot requires two measures (numerical fields). For this example, we'll use Sales and Profit.

From the "Tables" section on the left-hand Data pane, drag the Sales measure to the Columns shelf.
Next, drag the Profit measure to the Rows shelf.

Right now, your chart will show just a single mark. This is because Tableau has aggregated all sales and all profits into one dot. It’s showing you the total sales and total profit for the entire dataset combined. To create a scatter plot, we need to break this single point down into many individual marks.

Step 3: Add Detail to the View

To see a plot with many dots, you need to tell Tableau what each dot should represent. We want to see the sales and profit for each individual customer, so our "level of detail" is the customer.

From the Data pane, find the dimension Customer Name and drag it onto the Detail card in the Marks pane.

Instantly, the single dot will explode into a cloud of dots. Now, each mark on the chart represents a single customer, plotted according to their total sales and profit. You've officially created a scatter plot!

Quick Tip: Alternatively, you can go to the Analysis menu at the top and uncheck Aggregate Measures. This tells Tableau to plot every single row from your data source, which can be useful but often overwhelming. Adding a dimension like Customer Name to Detail is usually the better approach because it gives you control over the level of your analysis.

Making Your Scatter Plot More Insightful

A basic scatter plot is useful, but the real insights come when you add more context. Here’s how to enhance your visualization with a few extra steps.

Add a Trend Line

To make the correlation clearer, add a trend line. Go to the Analytics pane (next to the Data pane). Simply drag Trend Line from the list and drop it onto the Linear model option in the view. Tableau will instantly draw a line that shows the general trend in your data. In our Sales vs. Profit example, you’ll see the line going up, confirming a positive relationship.

Incorporate Color and Size

You can use color, size, and shape to layer in more information. What if you want to see if this Sales vs. Profit relationship differs by product category?

Drag the Category dimension to the Color card in the Marks pane.

Now, each dot is color-coded based on its category (Furniture, Office Supplies, or Technology). You can quickly see if one category is more or less profitable than others at similar sales levels.

Want to see which customers place the most orders?

Drag the Quantity measure to the Size card.

Larger dots will represent customers with higher order quantities, giving you another layer of visual information at a glance.

Customize Tooltips

Tooltips are the boxes of information that appear when you hover over a data point. By default, they show the fields you're using. You can make them more useful by clicking on the Tooltip card and adding more fields, like Region or State, providing deeper context without cluttering the chart.

The Challenge: Friction in the Analysis Process

As you can see, building and refining a scatter plot in Tableau involves a series of clicks, drags, and drops. While powerful, this manual process has its hurdles:

The Learning Curve: You need to know that a scatter plot requires two measures, that you must disaggregate the marks with a dimension, and where the Detail, Color, and Size cards are. For someone new, this isn't always intuitive.
Time and Iteration: Analysis is rarely a straight line. You might build the chart and realize Customer Name isn't the right level of detail. So you remove it, add Product Name, and see what that looks like. Then, you decide to color it by Region, then by Segment. Each of these steps takes time, breaking your analytical flow.
Cognitive Load: You're constantly translating a business question (“Are our bigger customers more profitable?”) into a series of technical steps within the tool. Your mental energy is spent on how to build the chart, not on what the chart is telling you.

The process works, but it isn't seamless. It’s like wanting to write a sentence and having to pick out each letter from a toolbox one by one. But what if you could just speak the sentence and have it appear?

A Faster Way: Creating Scatter Plots With AI

Instead of manually constructing your chart, AI allows you to get straight to the answer by simply describing what you want to see in plain English. This is at the heart of AI-driven analytics tools that use natural language to build visualizations for you.

Imagine just typing or speaking a prompt like:

“Show me a scatter plot of sales vs. profit for each customer.”

The AI handles the rest. It recognizes that "sales" and "profit" are your two measures, identifies "customer" as the level of detail, and instantly generates the exact scatter plot you built above, all without you touching a single shelf or Marks card.

How AI Streamlines Analysis

This natural language approach fundamentally changes how you interact with data:

Increased Speed: You can create a complex chart in seconds. A process that once took several minutes of thought and maneuvering around an interface now happens instantly.
Ultimate Accessibility: Anyone on your team, regardless of their data skills, can build reports and find insights. If you can ask a question, you can be a data analyst. This breaks down data silos and empowers more people to make data-informed decisions.
Fluid Exploration: Follow-up questions become effortless. After generating the initial plot, you can continue the conversation:
- “Now add a trend line.”
- “Color the points by product category.”
- “Ok, filter this to only show customers in California.”
- “Which customer has the highest profit?”

This conversational approach removes the friction, keeping you in your flow of analysis. You're no longer manipulating a tool, you're having a conversation with your data.

Final Thoughts

Scatter plots are an essential visualization for uncovering relationships between variables. Learning the mechanics of building them in tools like Tableau is a valuable skill that gives you a deeper understanding of your data's structure. However, the future of data analysis is faster, more conversational, and more accessible to everyone.

We designed Graphed to be that future. Instead of forcing you to learn the complex steps of a new tool, our AI data analyst builds dashboards and visualizations for you based on simple, natural language prompts. By connecting your data sources and just describing what you need, C-suite executives, junior marketers, and seasoned analysts alike can go from question to insight in seconds, focusing on strategy instead of struggling with software.