Solar Correlation Map
A radial layout that places a target variable at the center, with all other variables orbiting at distances proportional to their correlation strength.
// 01 — The chart
What it looks like
A solar correlation map with revenue at the center. Closer, darker nodes (like Ad spend) have stronger correlations; distant, lighter nodes (like Holidays) have weaker ones.
// 02 — Definition
What is a solar correlation map?
A solar correlation map is a radial visualization that positions a focal variable (the “sun”) at the center and arranges all other variables around it like orbiting planets. The distance from the center encodes correlation strength: strongly correlated variables sit close to the center, while weakly correlated ones orbit far away.
This layout is particularly effective for answering the question: “Which variables are most (and least) related to my target?” It provides an intuitive, ego-centric view of correlations that is easier for non-technical audiences to grasp than a full correlation matrix.
Additional visual channels — node size, color intensity, and connecting line thickness — can encode the sign of the correlation (positive vs. negative) and additional metadata.
Mental model: Think of a solar system — the sun (your target variable) is at the center, and planets (other variables) orbit at distances proportional to how “attracted” they are to the sun. Mercury (high r) is close; Neptune (low r) is far.
// 03 — Anatomy
Parts of a solar correlation map
// 04 — Usage
When to use it — and when not to
- You have one specific target variable and want to rank all others by relationship strength
- Presenting to non-technical stakeholders who find matrices intimidating
- You want an intuitive, visually engaging overview of variable importance
- Comparing which predictors are most relevant to a business KPI
- The ego-centric view (one focal variable) is more useful than all pairwise views
- You have 5–20 variables to position around the center
- You need to see correlations between all variables — use a correlogram instead
- You have only 2–3 variables — a scatter plot is more informative
- You need precise numeric comparisons — radial distance is harder to read than aligned axes
- Variables have important inter-relationships that should be shown (not just relationships to the center)
- Too many variables (40+) cause overlap and clutter in the radial layout
- You need to show non-linear or conditional relationships
// 05 — Reading guide
How to read a solar correlation map
Identify the central variable
The node at the center is the target variable — every other variable is measured in relation to it.
Check closest satellites
Variables nearest to the center have the strongest correlations with the target. These are your strongest predictors or associated factors.
Note the outermost nodes
Variables on the periphery have little or no linear relationship with the target. They can likely be deprioritized in analysis.
Check color or sign indicators
If the chart uses different colors for positive vs. negative correlations, note which variables move in the same direction as the target and which move opposite.
Read the orbit labels
Reference rings typically label correlation thresholds (e.g., r=0.8, 0.5). Use these for approximate quantification.
// 06 — Data format
What your data should look like
// Correlation to target
| variable | correlation_to_revenue |
|-----------|------------------------|
| ad_spend | 0.92 |
| traffic | 0.85 |
| weather | 0.15 |
// 07 — Construction
How to build one
Choose the target variable and compute its correlation with all other variables.
Map correlation strength to radial distance — r=1.0 is at the center, r=0.0 is at the outermost ring.
Distribute variables evenly around the circle at their computed distances to minimize label overlap.
Draw connecting lines from each satellite to the center node, with opacity or thickness proportional to r.
Add orbit reference rings with labeled r-values (e.g., 0.8, 0.5, 0.2) for quantitative anchoring.
// 08 — Pitfalls
Common mistakes
Overloading with too many nodes
With 30+ variables, labels overlap and the radial layout becomes a tangled mess. Filter to the top 15–20 most relevant variables.
No orbit reference rings
Without labeled concentric circles, readers can’t judge distances accurately. Always include them.
Ignoring inter-variable correlations
The solar map only shows relationships to the center variable. Two close satellites might be highly correlated with each other, which this layout doesn’t reveal.
Misleading distance mapping
Using an inverted scale (strong = far) or non-linear mapping confuses interpretation. Keep the convention: closer = stronger.
Treating it as causal
Correlation distance does not imply causation. A close satellite is not necessarily a driver — it could be a consequence or share a common cause.
// 09 — In the wild
Real-world examples
Business intelligence
Marketing teams use solar maps to show executives which campaign metrics (email opens, social clicks, paid impressions) are most strongly tied to revenue.
Healthcare
Researchers visualize which lifestyle factors (exercise, diet, sleep, stress) are most closely associated with a specific health outcome like blood pressure.
Feature selection
Data scientists use solar maps to quickly identify which features have the strongest linear relationships with the prediction target before model building.
// 10 — At a glance
Quick reference
Also known as
Radial correlation chart, ego network
Category
Correlation
Typical data
1 target + multiple numeric variables
Best for
Ranking variable associations to a target
Difficulty
Advanced
Key encoding
Distance from center = correlation strength
// 11 — Accessibility
Making it accessible
Label each node clearly with the variable name and r-value
Use shape or pattern differences for positive vs negative correlation nodes
Provide a sorted table alternative for screen readers
Ensure sufficient contrast between orbit rings and the background
Use tooltips on interactive versions for precise correlation values
// 12 — Variations
Common variations
Grouped solar map
Groups satellite variables by category (e.g., marketing, operations) using color-coded segments.
Multi-sun layout
Multiple target variables each get their own solar map, arranged side by side for comparison.
Interactive solar map
Clicking a satellite brings it to the center, rebuilding the map around the new target.
Partial correlation solar
Uses partial correlations (controlling for other variables) instead of raw Pearson r for more accurate relationships.
// 13 — FAQs
Frequently asked questions
What is a solar correlation map?+
A solar correlation map is a radial visualization that positions a focal variable (the "sun") at the center and arranges all other variables around it like orbiting planets. The distance from the center encodes correlation strength: strongly correlated variables sit close to the center, while weakly correlated ones orbit far away.
When should you use a solar correlation map?+
Use a solar correlation map when you have one specific target variable and want to rank all others by relationship strength. It also works well when presenting to non-technical stakeholders who find matrices intimidating, and when you want an intuitive, visually engaging overview of variable importance.
When should you avoid a solar correlation map?+
Avoid a solar correlation map when you need to see correlations between all variables — use a correlogram instead. It is also a poor fit when you have only 2–3 variables — a scatter plot is more informative, or when you need precise numeric comparisons — radial distance is harder to read than aligned axes.
What data do you need to make a solar correlation map?+
// Correlation to target
How is a solar correlation map different from a correlogram?+
Both a solar correlation map and a correlogram can look similar at first glance, but they answer different questions. Reach for a solar correlation map when the comparisons and patterns it was designed to reveal match what you need to communicate, and choose a correlogram when its particular strengths better fit your data and audience.
What is another name for a solar correlation map?+
Solar Correlation Map is also known as Radial correlation chart, ego network. The name varies between fields, but the visualisation technique is the same.
What size of dataset works best for a solar correlation map?+
Solar Correlation Map works best for Ranking variable associations to a target. Outside that range the chart either looks empty or becomes too cluttered to read clearly.
Are solar correlation maps accessible to screen readers?+
Yes — a solar correlation map can be made accessible to screen readers by pairing it with a clear text summary of the key insight, ensuring color choices meet WCAG contrast guidelines, adding descriptive alt text or aria-label to the SVG, and offering the underlying data as an HTML table fallback for assistive technologies.