How to Make an X Y Graph in Looker with AI

Creating an X/Y graph, or scatter plot, is one of the fastest ways to see if a relationship exists between two different metrics. This article will show you how to build one step-by-step in Looker Studio (formerly Google Data Studio), covering both the traditional click-by-click method and the newer, faster AI-powered approach.

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 an X/Y Graph and Why Should You Use It?

An X/Y graph, more formally known as a scatter plot, places dots on a graph to represent the relationship between two numerical variables. One variable sits on the horizontal X-axis, and the other sits on the vertical Y-axis. Each dot on the chart represents a single data point, showing its value for both variables simultaneously.

The main reason to use an X/Y graph is to spot correlations. Are two things related? If one goes up, does the other go up too? Or down? Or does nothing happen at all?

Think of it with these common business examples:

• Marketing Spend vs. Sales: Does your revenue (Y-axis) increase as you spend more on ads (X-axis)? A scatter plot quickly visualizes your return on ad spend (ROAS).
• Website Traffic vs. Conversions: Do a higher number of user sessions (X-axis) on a landing page lead to more sign-ups (Y-axis)?
• Sales Activity vs. Deals Closed: Do sales reps who make more calls (X-axis) tend to close more deals (Y-axis)?

When you look at a scatter plot, you're trying to identify a pattern:

• Positive Correlation: The dots form a rough line that goes up from left to right. This suggests that as your X variable increases, your Y variable tends to increase, too (e.g., more ad spend, more sales).
• Negative Correlation: The dots form a line that goes down from left to right. As X increases, Y tends to decrease (e.g., a higher website bounce rate is correlated with lower time on page).
• No Correlation: The dots are scattered all over the chart like a random shotgun blast. This suggests there’s no clear relationship between the two variables.

The “Classic" Way: How to Create an X/Y Graph Manually in Looker Studio

The manual method is all about navigating the menus and side panels in Looker Studio. It gives you full control but requires you to know where everything is. Let's walk through it using an example: comparing user sessions vs. conversions for various landing pages in Google Analytics 4.

Step 1: Create a New Report and Connect Your Data

First, log in to Looker Studio and start a new blank report. The first thing it will ask you to do is connect a data source. Find and select the source you want to use, such as "Google Analytics." Authorize the connection and choose your GA4 property.

If your data is in a spreadsheet, you can select "Google Sheets" and connect to the proper file.

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

Step 2: Add a Scatter Chart to Your Canvas

Once your report canvas is ready, go to the toolbar at the top and click Add a chart. Scroll down through the options until you find the "Scatter" category and select the first one, the basic scatter chart.

Click anywhere on the report canvas to drop the chart in. Looker will populate it with default dimensions and metrics, which we’ll need to change.

Step 3: Configure Your Metrics and Dimensions

With the new chart selected, a "Properties" panel will appear on the right side of your screen. This is where you tell the chart what data to display. Under the "Setup" tab, you’ll see fields for Dimensions, Metric X, and Metric Y.

• Dimension: This tells Looker what each dot on the graph should represent. Since we want to analyze landing page performance, we’ll drag the Landing page + query string dimension here. You could also use Campaign name, Product SKU, or Country.
• Metric X: This is the data for your horizontal axis. We want to see how sessions affect conversions, so let's drag the Sessions metric here.
• Metric Y: This is the data for your vertical axis. Here, we’ll drag the Conversions metric.

As you add these, the chart on your canvas will update in real-time. You should now see a collection of dots, where each dot is a unique landing page, and its position is determined by its number of sessions (X) and conversions (Y).

Step 4: Style Your Graph for Clarity

An unstyled graph can be tough to read. In the "Properties" panel, switch from the "Setup" tab to the "Style" tab.

A few crucial tweaks can make a huge difference:

• Add a Trendline: This is maybe the most important styling feature for a scatter plot. Under the "General" section (you may need to scroll down), find the "Trendline" option and select "Linear." This draws a line of best fit through your data points, making it instantly clear if the correlation is positive, negative, or flat.
• Adjust Points: You can change the color and size of the dots to make them easier to see.
• Gridlines and Axes: Clean up the background by changing the grid color (or removing it) and add titles to your X and Y axes (e.g., "Total Sessions" and "Total Conversions") so everyone knows what they’re looking at.

The Faster Way: Building Your X/Y Graph with Looker Studio's AI

Manually creating charts isn't difficult once you know the steps, but it involves hunting through menus and remembering what goes where. Recently, Looker Studio has added AI capabilities that let you generate charts just by describing what you want in plain English. This can cut the creation process down to seconds.

This feature is available in Looker through an AI chat box or a "Generate with a prompt" button. Depending on your version and organization's settings, it might be labeled slightly differently - look for something that involves generative AI or building reports via prompts.

Step 1: Open the AI Prompt Interface

Instead of clicking "Add a chart," find the AI icon or button in the Looker Studio interface, frequently located in the toolbar. This will open a chat or prompt window that allows you to give instructions.

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 2: Write a Clear and Specific Prompt

The quality of the AI's output depends entirely on the quality of your prompt. You need to tell it exactly what you want it to build. A vague prompt like "show traffic and leads" is likely not effective.

A good prompt should include:

1. The chart type ("scatter plot" or "X/Y graph")
2. The primary metric for the X-axis
3. The second metric for the Y-axis
4. The dimension to break down the data by
5. The data source (if multiple are connected)

Let's use our same example. Here is a strong prompt:

Create a scatter plot comparing sessions vs conversions by landing page from our Google Analytics 4 data.

Or, for another example:

Build an xy graph showing a comparison of Ad Cost and Purchases for each Campaign Name from our Google Ads account.

Step 3: Insert and Refine the Chart

After you submit your prompt, the AI will process it and generate a chart that best matches your request. It will automatically populate the dimensions and metrics in the properties panel for you.

Click a button to add this chart to your report canvas. The best part is that this AI-generated chart is not a static image, it's a fully editable Looker Studio chart. You can now use the same "Style" panel we used in the manual method to add a trendline, change colors, or adjust filters. The AI handles the initial setup and configuration, which is often the most time-consuming part, letting you jump straight to analysis.

Tips for Getting More From Your X/Y Graphs

Building the chart is just the first step. Here's how to interpret what you see and find meaningful insights.

Always Add a Trendline

It's worth repeating: a trendline is your best friend when working with a scatter plot. The direction of the line immediately tells you the nature of the relationship (positive or negative), and its steepness gives you a general idea of the strength of that relationship. A nearly flat line suggests a very weak correlation, whereas a steep line indicates a much stronger connection between the variables.

Investigate the Outliers

Outliers are the dots that live far away from the main cluster and the trendline. Don't ignore them! They often represent the most interesting stories in your data.

Did one particular landing page get a ton of traffic but almost zero conversions? Why? Find that page and investigate. Was there a technical error? Or is the messaging simply not working? On the other end, is there a page with few sessions but a sky-high conversion rate? That's a huge opportunity. That page is a high-performer that you should learn from and try to send more traffic to.

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

Remember: Correlation Is Not Causation

This is a fundamental rule of data analysis. Just because two things move together doesn't mean one causes the other. A scatter plot might show a strong positive correlation between ad spend and revenue, which is a great sign. However, there could be other factors at play. Maybe you ran a huge PR campaign at the same time, or a seasonal event occurred. The correlation is a signal to dig deeper, not a final conclusion.

Segment Your Data for Deeper Insights

A high-level X/Y graph tells a general story. A segmented graph tells you where the real opportunities are. Instead of plotting all your sessions vs. conversions, try applying a filter or creating separate charts for different segments:

• By Channel: Does sessions vs. conversions look different for Organic Search vs. Paid Social?
• By Device: Does mobile traffic have a stronger (or weaker) relationship with conversions compared to desktop traffic?
• By Region: Do users from the US show a stronger correlation between sessions and purchases than users from Canada?

Segmentation often reveals hidden patterns and lets you make more precise, effective business decisions.

Final Thoughts

Whether you're manually configuring every setting or using natural language prompts, creating an X/Y graph in Looker Studio is a powerful way to quickly spot relationships in your data. It's a fundamental chart for any marketer, analyst, or business owner who wants to move beyond surface-level metrics and understand what drivers are truly impacting their goals.

Moving from manual clicking to simple prompts is a huge leap in efficiency. That's why we're building Graphed to take this idea even further. While Looker’s AI helps you build a single chart, we made it so you can ask conversational questions to build entire real-time dashboards that pull from all your scattered data sources. You can just ask things like, "Show me how our Facebook Ads spend correlates with our Shopify sales this quarter," and get an answer in seconds, without ever wrangling different platforms yourself.