Multi-dimensionalIntermediate

Radar Chart

A chart that plots multiple variables on axes radiating from a center point, forming a polygon — revealing the shape of your data at a glance.

// 01 — The chart

What it looks like

Example — Product evaluation across six criteriaQ2 2025
PerformanceUsabilityReliabilityCostSupportFeaturesProduct AProduct B

A radar chart comparing two products across six evaluation criteria. The filled polygon (Product A) shows stronger performance and usability.

// 02 — Definition

What is a radar chart?

A radar chart (also called a spider chart, web chart, or star plot) displays multivariate data on a two-dimensional chart with three or more quantitative variables. Each variable gets its own axis, and all axes share the same central origin point, radiating outward like spokes of a wheel.

Data points on each axis are connected to form a closed polygon. The shape of this polygon is the key insight — it reveals the overall profile of the entity being measured. A perfectly balanced entity produces a regular polygon; an unbalanced one produces a distinctive, lopsided shape.

Radar charts are particularly powerful for comparing profiles. When you overlay two or more polygons, you can instantly see where one entity excels or falls short relative to another. They turn abstract multi-dimensional comparisons into visual patterns your brain can process at a glance.

Origin: The radar chart was first developed by Georg von Mayr in 1877 for his work on Bavarian population statistics. It gained popularity in the mid-20th century when the Japanese industrial sector adopted it for quality control processes, giving rise to its alternative name “spider diagram.”

// 03 — Anatomy

Parts of a radar chart

ABCDE
A — Center (origin): The common starting point for all axes, representing zero or the minimum value
B — Spokes (axes): Lines radiating from the center, each representing one variable being measured
C — Grid rings: Concentric polygons showing equal value intervals, helping the eye gauge data point positions
D — Data polygon: The filled or outlined shape formed by connecting all data points — its silhouette is the key visual
E — Data points: Individual dots on each axis showing the exact value for that variable

// 04 — Usage

When to use it — and when not to

✓Use a radar chart when…
  • Comparing the overall profile of 2–3 entities across multiple dimensions
  • You have 3–8 variables that share a comparable scale
  • The shape of the data matters more than individual values
  • Showing strengths and weaknesses in a performance review or product evaluation
  • You want to highlight balance or imbalance across categories
  • Presenting skill assessments, competency maps, or feature comparisons
×Avoid a radar chart when…
  • You need precise value comparison — bar charts are far more accurate for exact readings
  • You have more than 3 overlapping data series, which becomes an unreadable mess
  • Variables are on wildly different scales (0–1 vs. 0–10,000)
  • You have more than 10 axes — the polygon becomes too complex to interpret
  • You want to show change over time — use a line chart instead
  • Your audience isn’t familiar with radar charts — they require more effort to read than bars

// 05 — Reading guide

How to read a radar chart

Follow these steps whenever you encounter a radar chart in the wild.

1

Identify the axes and their labels

Start by reading each spoke label to understand what variables are being measured. Check whether there’s a scale on any spoke — all axes should share the same scale range for the chart to be meaningful.

2

Look at the overall shape

Before reading individual values, step back and observe the polygon’s silhouette. A large, regular shape suggests strong and balanced performance. A small or spiky shape indicates weaknesses or unevenness.

3

Identify peaks and valleys

The points farthest from the center are the highest-scoring variables. The points closest to the center are the weakest areas. These extremes are usually the most important story in the data.

4

Compare overlapping polygons

If multiple entities are plotted, compare their shapes. Where does one polygon extend beyond another? Where does it fall short? The overlap and divergence patterns reveal relative strengths and weaknesses.

5

Read individual values last

Once you understand the overall pattern, check specific axis values for precision. Use the gridlines to estimate numeric values for any data points that stand out from your shape analysis.

// 06 — Common mistakes

Mistakes to watch out for

Too many variables (axes)

Once you exceed 8–10 axes, the chart becomes a dense web that’s nearly impossible to interpret. The polygon shape loses meaning because the eye can’t distinguish individual data points. Reduce variables to the most important 5–7, or consider parallel coordinates or small multiples instead.

Overlapping too many data series

Three overlapping polygons is the practical maximum. Beyond that, colors blend, edges overlap, and comparison becomes impossible. If you need to compare more than three entities, use small multiples — one radar chart per entity with the same scale.

Axes with different scales

If one axis runs from 0 to 100 and another from 0 to 10,000, the chart is misleading. All axes must share the same normalized scale (usually 0–1 or 0–100%). Without normalization, the polygon shape communicates false information about relative performance.

Axis ordering affects perception

The order in which axes are arranged changes the polygon’s shape and visual impression. Adjacent high values create a “peak” while adjacent low values create a “valley.” Be deliberate about axis ordering and don’t rearrange axes to make a polygon look better.

Area misinterpretation

Because radar charts use a radial layout, the visual area of the polygon is not proportional to the sum of values. A polygon that appears twice as large doesn’t represent twice the total. Don’t compare polygon areas — compare individual axis positions instead.

// 07 — Real-world examples

Where you’ll see radar charts used

01

Gaming: Player attribute profiles

Sports video games like FIFA and NBA 2K use radar charts to display player attributes — speed, shooting, passing, defense, physical. The polygon shape instantly communicates whether a player is a well-rounded all-rounder or a specialist with extreme peaks and valleys.

Gaming & Sports
02

Business: Competitive product analysis

Product teams overlay radar charts for competing products across dimensions like price, features, reliability, support, and ecosystem. The visual comparison reveals each product’s unique positioning and helps identify market gaps.

Product Strategy
03

HR: Employee competency assessments

Human resources departments use radar charts to visualize employee skill profiles across communication, technical ability, leadership, teamwork, and problem-solving. Comparing current skills to role requirements highlights training needs and career development paths.

Human Resources

// 08 — At a glance

Quick reference

Also known asSpider chart, web chart, star plot, Kiviat diagram
First used byGeorg von Mayr, 1877
Best forComparing multi-dimensional profiles of 2–3 entities
Data types3–8 quantitative variables on a common or normalized scale
Recommended axes5–7 for clarity
Max data series3 overlapping polygons. Use small multiples beyond that.
Common toolsExcel, Chart.js, D3.js, Matplotlib, Plotly, Tableau
Common mistakesToo many axes, mixed scales, too many overlapping series, area misinterpretation

// 09 — Variations

Types of radar charts

The basic radar chart has several variants adapted for specific use cases.

Filled radar chart

Uses a semi-transparent fill to emphasize the area covered by the data polygon. Best for single-entity profiles where overall coverage matters.

Comparative radar chart

Overlays two or more data polygons to directly compare profiles. Divergences between the shapes reveal relative strengths and weaknesses.

Circular radar chart

Uses circular gridlines instead of polygonal ones. The data polygon remains, but the concentric circles make it easier to judge distance from center.

Nightingale rose chart

Uses wedges of varying radius instead of connected points. Each sector’s length (not area) encodes the data value. Invented by Florence Nightingale in 1858.

// 10 — FAQs

Frequently asked questions

What is a radar chart?+

A radar chart (also called a spider chart, web chart, or star plot) displays multivariate data on a two-dimensional chart with three or more quantitative variables. Each variable gets its own axis, and all axes share the same central origin point, radiating outward like spokes of a wheel.

When should you use a radar chart?+

Use a radar chart when comparing the overall profile of 2–3 entities across multiple dimensions. It also works well when you have 3–8 variables that share a comparable scale, and when the shape of the data matters more than individual values.

When should you avoid a radar chart?+

Avoid a radar chart when you need precise value comparison — bar charts are far more accurate for exact readings. It is also a poor fit when you have more than 3 overlapping data series, which becomes an unreadable mess, or when variables are on wildly different scales (0–1 vs. 0–10,000).

How is a radar chart different from a bar chart?+

Both a radar chart and a bar chart can look similar at first glance, but they answer different questions. Reach for a radar chart when the comparisons and patterns it was designed to reveal match what you need to communicate, and choose a bar chart when its particular strengths better fit your data and audience.

Is a radar chart suitable for dashboards?+

Yes — a radar chart can work well in dashboards as long as the panel is large enough for readers to perceive the encoded values, has a clear title, and includes the legend or axis labels needed to interpret it.

What category of chart is a radar chart?+

Radar Chart belongs to the Multi-dimensional family of charts. Charts in that family are designed to answer the same kind of question, so they often work as alternatives when one doesn't quite fit your data.