Free Scatter Plot & XY Chart Maker

Scatter plot, scatter chart, and XY chart describe the same visualization: a 2D grid with a point plotted for each pair of numerical values. The different names come from different contexts. Statistics courses and research writing tend to say "scatter plot". Excel and business dashboards tend to say "scatter chart". Engineering and finance often say "XY chart" because both axes are continuous numerical (X and Y) rather than categorical. The visualization is identical across all three names.

Our tool produces all three from the same widget. Drop data in, style the points, label the axes, export. The vocabulary your audience uses is whatever they encountered first.

Use a scatter plot when you have two continuous numerical variables and you want to see the relationship between them. The shape of the cloud tells the story: a tight cloud sloping up = strong positive correlation; a loose cloud with no slope = weak or no correlation; two separated clouds = distinct subgroups in the data that may need separate analysis.

Scatter plots are the starting point for statistical analysis. Before computing a correlation coefficient or fitting a regression model, the plot tells you whether the relationship is linear, nonlinear (curved), clustered, or nonexistent. Skipping the plot and jumping straight to numbers often misses patterns that are obvious visually.

Skip a scatter plot when one axis is categorical (product names, regions, survey options). For one numerical and one categorical variable, use a bar chart at bar chart maker or a box plot. For one numerical variable over time, use a line chart at line chart maker.

Shape of the cloud: a cloud that trends strongly upward as you move right means X and Y are positively correlated. Downward-trending means negatively correlated. Flat or scattered means weak or no correlation. The tighter the cloud hugs a line, the stronger the relationship.

Outliers: points that sit far from the main cloud deserve attention. Sometimes they are data entry errors worth removing; sometimes they are the most interesting observations in the dataset (unusually successful customers, unusually fast sales cycles, unusually extreme experimental results).

Clusters: if the cloud is actually two or more separate clouds, the data has subgroups. A single trend line across mixed clusters is misleading — fit separate trend lines per cluster, or investigate what makes the subgroups different.

Density: at the center of a scatter, overlapping points can hide density. Increasing point transparency (alpha) reveals how many points overlap. For very large datasets, a heatmap or hexbin chart reads better than a raw scatter.