Need a scatter plot in Excel and not sure where to start? You're in the right place. A scatter plot, sometimes called an XY chart, plots two numeric variables against each other so you can spot trends, clusters, and outliers in seconds. It's one of the most useful chart types in Excel, and once you've made one, you'll keep coming back to it.
In this guide, you'll learn how to make a scatter plot in Excel from scratch — selecting data, inserting the chart, customizing axes, adding labels, and styling everything for clarity. Whether you're a student learning data analysis, a marketer studying campaign performance, or an analyst working with sales figures, the steps are the same. Excel handles the math; you handle the story.
We'll cover Excel 2016, 2019, 2021, and Microsoft 365. The menus look almost identical across versions, so don't worry if your screen doesn't match a screenshot exactly. The logic is what matters. By the end, you'll be making clean, professional scatter plots without breaking a sweat — and you'll know when to use them instead of a line chart or bar chart.
Quick heads-up: scatter plots need numeric data on both axes. If one of your variables is a date or category, you probably want a line chart or column chart instead. Keep that in mind before we start clicking.
Before we click anything, a quick sanity check on your data. Excel needs your X values (independent variable) and Y values (dependent variable) in two adjacent columns, with X on the left. Headers are fine — actually, they're encouraged because Excel uses them for the legend and axis titles. Don't leave blank rows between data points; they confuse the chart engine and sometimes break the trendline calculations later.
Here's what a clean dataset looks like: column A has "Hours Studied" with values from 1 to 10, and column B has "Test Score" with values from 50 to 98. Each row is one student. That's it. Two columns, headers, no merged cells, no blank rows. If your data lives in a table (Insert → Table), even better — charts auto-expand when you add new rows.
Use a scatter plot when both variables are numeric and you want to see if they're related. Classic examples: hours studied vs. test score, ad spend vs. revenue, temperature vs. ice cream sales, age vs. blood pressure. If one axis would be categories (red/blue/green) or time labels (Jan/Feb/Mar), use a bar or line chart instead.
Let's walk through the actual clicks. Open your workbook, find your two columns of numeric data, and follow along. We'll start with the simplest scatter plot — just dots on a grid — then layer on customization.
Step 1: Select your data. Click the top-left cell of your range and drag to the bottom-right. Include the headers. If your data is in A1:B25, select all of it. Keyboard shortcut: click A1, then press Ctrl+Shift+End to grab everything down and right.
Step 2: Open the Insert tab. Up in the ribbon, click Insert. Look for the Charts group in the middle. You'll see a small icon that looks like dots on a grid — that's the Scatter button. Hover over it and the tooltip says "Insert Scatter (X, Y) or Bubble Chart."
Step 3: Pick a subtype. Click the Scatter icon and a dropdown appears with five options: Scatter (dots only), Scatter with Smooth Lines and Markers, Scatter with Smooth Lines, Scatter with Straight Lines and Markers, and Scatter with Straight Lines. For most uses, the first option (just dots) is the right choice. Click it.
Step 4: Excel draws the chart. A scatter plot appears on your worksheet, usually floating over your data. You can click and drag it anywhere. Use the corner handles to resize. Don't worry if it looks plain — we'll style it next.
The classic. Just markers showing each data point. Best for showing correlation without implying a sequence between points.
Connects points with a curved line. Useful when your X values represent a continuous progression (like time or temperature steps).
Curved line only, no dots. Cleaner look for presentations when you don't need to highlight individual data points.
Connects points with straight segments. Good for engineering or scientific data where exact values matter at each point.
Just the connecting lines. Use when the trend matters more than individual readings — think rough sketches of relationships.
The default scatter chart is functional but bland. Let's make it readable. Click anywhere inside the chart to activate it. You'll see three small icons appear on the right side: a plus sign (Chart Elements), a paintbrush (Chart Styles), and a funnel (Chart Filters). The plus sign is where you'll spend most of your time.
Click the plus icon. You'll see checkboxes for Axes, Axis Titles, Chart Title, Data Labels, Error Bars, Gridlines, Legend, and Trendline. At minimum, turn on Axis Titles and Chart Title. Leave Gridlines on for now — they help readers estimate values. The Legend is optional; for a single-series scatter plot, you can usually turn it off because the chart title says everything.
For Chart Title, click the text that appears at the top of the chart and type something descriptive. "Hours Studied vs. Test Score" beats "Chart 1" every time. For Axis Titles, click each placeholder and label them with the variable name plus units — "Hours Studied (h)" on the X-axis, "Test Score (out of 100)" on the Y-axis.
Data labels show the actual value next to each dot. They're useful when you have fewer than 15 points and each one matters. Click the plus icon → check Data Labels. By default, Excel shows the Y value. To show X and Y together, right-click a label → Format Data Labels → check both "X Value" and "Y Value." To label points with names from a third column (like student names), use the "Value From Cells" option in newer versions of Excel.
Trendlines are where scatter plots really earn their keep. They show the underlying relationship between your two variables. Right-click any data point and choose "Add Trendline." In the panel that opens, pick Linear if you expect a straight relationship, Polynomial (order 2 or 3) if you see a curve, or Exponential for compounding growth patterns. Always check "Display R-squared value on chart" — it tells you how well the trendline fits. An R² above 0.7 is a decent fit; above 0.9 is strong; below 0.3 means your variables probably aren't very correlated.
One trap to avoid: don't extend your trendline far beyond your data range using the Forecast option unless you have good reason. Extrapolation can be misleading, especially with polynomial fits that go wild at the edges.
Even simple scatter plots can misbehave. If your chart appears completely empty after insertion, double-check that you actually selected numeric values and not just headers. Excel sometimes grabs only the header row if you click once on a cell instead of dragging across the range. Re-select the full data block, including all rows, and re-insert.
If the dots cluster on top of each other and you can't tell how many points are in each spot, you're dealing with overplotting. Bump marker transparency: right-click a point → Format Data Series → Marker → Fill → set transparency to around 50%. Now overlapping dots show darker, giving you a rough density read. For serious overplotting (thousands of points), jitter your data slightly with a RAND() column or switch to a 2D histogram in Power Query.
If your X-axis shows weird category labels like "1, 2, 3, 4" instead of your actual X values, the chart got inserted as a Line type by mistake. Click the chart, go to Chart Design → Change Chart Type, and pick Scatter (X, Y) from the left panel. Excel will recompute axes using the real numeric distances.
Finally, if your trendline looks bizarre — shooting off to infinity or wiggling all over — you probably picked the wrong fit. Polynomial order 4 or higher overfits almost any data; drop back to order 2 or switch to Linear. Always check R-squared; if it's below 0.5, the trendline is more wishful thinking than insight.
What if you want to compare two groups on the same chart — say, test scores for Class A and Class B? You can plot multiple data series on a single scatter plot. The trick is structuring your data correctly. Set up four columns: Class A X, Class A Y, Class B X, Class B Y. Then insert a scatter chart from the Class A pair, and use the Select Data dialog to add Class B as a second series.
Right-click the chart → Select Data → Add. In the dialog, give the series a name ("Class B"), point Series X values to the Class B X column, and Series Y values to the Class B Y column. Click OK twice. Excel adds the second series in a different color automatically. Now both groups appear on the same plot, and the legend tells you which is which.
This technique scales: you can add three, four, or even ten series the same way. Just don't get carried away — more than five series and the chart turns into confetti. If you need to compare many groups, consider a small multiples approach (one chart per group) instead.
Let's look at three scenarios where a scatter plot earns its keep. First: a marketing analyst tracks ad spend (X) against sales revenue (Y) across 30 campaigns. The scatter plot reveals a positive linear trend up to about $5,000 in spend, then flattens — classic diminishing returns. The trendline equation gives the team a rough formula to predict revenue from spend, and the breakpoint tells them where to cap budgets.
Second: a school counselor plots study hours (X) against final grades (Y) for 50 students. The scatter shows a strong positive correlation (R² = 0.78), but a cluster of points in the upper-left corner reveals a handful of students who scored well despite minimal studying — outliers worth a closer look. Maybe they had prior knowledge, or maybe they cheated. Either way, the chart surfaced something the average alone would have hidden.
Third: a fitness app developer plots app usage minutes (X) against weight loss in pounds (Y) for beta testers. The scatter shows almost no correlation (R² = 0.12). That's a finding too — it tells the team that app usage alone doesn't predict outcomes, and they need to investigate which features inside the app actually drive results. Sometimes the most valuable scatter plot is the one that shows no relationship.
Once your chart looks good, you'll probably want it outside Excel. Right-click the chart and choose "Save as Picture" — PNG keeps it sharp, JPG keeps file size small. For reports, copy the chart (Ctrl+C), then paste-special into Word as a picture so it doesn't break when colleagues open the file without your data.
To embed the chart in PowerPoint while keeping it editable, copy the chart, switch to PowerPoint, and paste with the "Use Destination Theme & Embed Workbook" option. The chart will live in your slide and can still be edited if you double-click it. If you want it to update when the Excel data changes, use "Keep Source Formatting & Link Data" instead — just keep both files together.
For web sharing, the cleanest path is to save as PNG at a generous resolution (Excel exports at screen resolution by default, which often looks blurry). One workaround: hold Shift while clicking File → Save As to access additional export options in some versions, or simply increase your zoom level before exporting so the rendered image is higher resolution.
Once you're comfortable with standard scatter plots, a few advanced techniques will set your work apart. Color-coding by category is the first big one. Say you have three product lines on one chart — Premium, Standard, Budget. Instead of one giant blob, you want each line to have its own color. The trick: split your Y values into three columns (one per category, leaving blanks where data doesn't apply) and add them as separate series. Now each category gets its own marker color and legend entry.
Conditional formatting on markers is another favorite. You can color individual points based on a third variable — say, highlight all data points where revenue exceeded a threshold in red, and leave the rest blue. There's no direct UI for this in Excel; you have to manually split the series or use VBA. But it's powerful for presentations: one chart, three or four insight layers.
Quadrant analysis is a classic strategic tool. Draw vertical and horizontal reference lines at the median X and median Y values, splitting your plot into four quadrants. Top-right = high on both variables (your stars), bottom-left = low on both (your dogs), and the off-diagonal quadrants represent mismatches worth investigating. You can add these reference lines by inserting a second "helper" series with just two points each, formatted as a thin line.
Scatter plots are a foundation skill in Excel and data analysis generally. Once you've made a handful, you'll instinctively know when to reach for one. A few last tips: always look at the chart with fresh eyes before sharing it — if a colleague glanced at it for three seconds, would they understand the story? If not, the title, axis labels, or trendline need more work.
Don't fall into the trap of decoration. Removing chart junk (background fills, heavy gridlines, 3D effects, unnecessary legends) almost always improves a scatter plot. Edward Tufte's idea of "data-ink ratio" applies: maximize the ink that shows data, minimize everything else. White backgrounds, thin gridlines, dark markers, and a clean title beat colorful gradients every time.
If you find yourself making scatter plots constantly, learn keyboard shortcuts. Alt+F1 inserts a default chart from selected data instantly. F11 creates a chart on a new sheet. Ctrl+1 with a chart selected opens the Format pane. These shortcuts compound over months — you'll save hours a year just on chart creation.
Ready to push further? The next chart types worth learning are bubble charts (scatter plot with a third variable as dot size), combo charts (mixing column and line on one chart), and box-and-whisker plots for distribution comparisons. Excel can handle all of them, and the muscle memory you built making scatter plots transfers directly. Good luck — and remember, the best scatter plot is the one that answers a real question.