Scatter

A scatter plot shows relationships between two numeric fields. Use it to spot correlations, clusters, and outliers.

A scatter plot shows individual data points based on an X and Y value. Use it to spot correlations, clusters, and outliers.

Scatter plot

Configure a scatter plot

Use the Plot panel to choose axes, grouping, and display settings.

Scatter plot options

Use these options to control how the scatter plot renders.

Option	What it controls	Typical values / notes
X-axis	Field used for the X axis.	Use a numeric field (for example latency, payload_size). You can also use timestamp for time-based scatter plots.
Y-axis	Field used for the Y axis.	Use a numeric field (for example error_rate, cpu_usage, value).
Group by	Split points into multiple series.	For example pod colors points by pod. Leave empty for a single series.
Y-axis label	Display label for the Y axis.	Cosmetic only. Example: “CPU Usage (%)”.
Y-axis scale	How values are scaled on the Y axis.	Linear for most data. Logarithmic for wide ranges.
Reference line	A horizontal threshold line.	Example: 80 for an SLA/limit line.
Legend	Show/hide legend.	Turn off for a single series or tight layouts.
Calculations	Summary overlays on the plotted data.	Min, Max, Avg, Count, Sum, P50, P90, P95, P99.
Custom color	Set fixed colors for series.	Helps keep colors consistent across dashboards.
Apply	Applies changes to the widget.	Changes do not render until you click Apply.

When to use a scatter plot

Use a scatter plot when you want to:

• Understand the relationship between two numeric metrics

• Spot correlation (positive/negative) or lack of correlation

• Identify clusters (groups of similar behavior)

• Find outliers (rare or extreme points)

Typical use cases:

• Latency vs request size

• CPU usage vs memory usage

• Error rate vs throughput

• Cost vs usage (per service/team)

When not to use a scatter plot

Avoid a scatter plot when:

• One axis is not numeric (use Bar/Table instead)

• You mainly want a trend over time (use Line/Area)

• You need exact record-level inspection (use Table/List/Details)