Choose Two Variables to Explore
The studio will auto-detect whether each variable is continuous or categorical and recommend the best chart types.
Data Visualization Studio
Descriptive Analytics
Bivariate Exploration
Explore bivariate relationships through the right visualization. Upload a dataset, pick two variables, and instantly see which chart types reveal the most about your data — with rich interpretation to help you understand what the patterns mean for real marketing decisions.
OVERVIEW & WHAT THIS TOOL SOLVES
The exact same marketing dataset can tell two contradictory stories depending on which chart you put in the slide deck: a bar chart of group averages can hide a polarized bimodal customer base, while a dense scatter plot can bury a trend under overplotted points. The Data Visualization Studio helps you match the right bivariate chart to your variable types—and compare multiple chart views side-by-side so you see both what a visualization reveals and what it hides.
🎯 Questions It Answers
- Which chart family best fits our variable pair (Continuous × Continuous, Categorical × Continuous, or Categorical × Categorical)?
- What hidden distribution features (bimodality, skewness, outliers, unequal group sizes) appear when we switch from a Bar Chart to a Box, Violin, or Strip Plot?
- How does relationship direction, clustering, or group composition shift when we facet or compare charts side-by-side?
📂 What Data You Need
- Marketing Dataset: Preset marketing cases (ad spend, store formats, customer segments) or uploaded CSV
- Two Focal Variables: Select an X variable and a Y variable (continuous numeric or categorical label)
- Chart Type & Comparison View: Switch between recommended chart types or open the side-by-side comparison panel
💡 What You'll Discover
- Smart Chart Selector: Variable-type-aware recommendations (Scatter, 2D Heatmap, Box, Violin, Strip, Grouped/Stacked Bar)
- Side-by-Side Chart Comparison: Direct visual contrast showing what summary bars conceal vs. distribution-aware plots
- Guided Narrative & Summary Table: Automatic group/bivariate summary statistics paired with chart-reading guidance
🔍 Key Nuances & Common Traps to Watch For
- Bar Charts of Means Hide the Distribution Shape: Two store formats can have the exact same $50 average basket while Format A is tightly clustered at $48–$52 and Format B is split between $15 convenience buyers and $120 bulk buyers. Always inspect a Box, Violin, or Strip plot before relying on a bar chart of averages.
- Counts vs. 100% Proportions in Categorical Charts: Grouped bar charts show raw volume differences across segments, whereas 100% Stacked Bar charts normalize group sizes so you can compare conditional conversion or preference shares fairly.
- Overplotting in Large Scatter Plots: When thousands of customers share similar values, scatter dots stack on top of one another and hide where the bulk of the density sits—switch to a 2D Binned Heatmap or add transparency to see the true concentration.
QUICK START: Choose Your Path
DATA SOURCE
📚
Use a Case Study
Load a preset marketing scenario to explore with different chart types.
📤
Upload Your Data
Drag & Drop raw data file (.csv, .tsv, .txt, .xls, .xlsx)
Up to 5,000 rows. Each column becomes a variable you can explore.
No file uploaded.
DATA INPUT
Select a dataset above to preview rows and columns.
VARIABLE SELECTION
CHART TYPE
COMPARE CHART TYPES
See how the same data looks across different chart types — and learn what each one reveals vs. hides.
SUMMARY TABLE
Summary table will appear after visualization.
WHAT CHARTS CAN HIDE
💡 The Danger of Aggregation
A bar chart of group means looks clean and decisive. But it hides everything that matters most:
- Distribution shape — Are values normally distributed, or bimodal? A bar chart can't tell you.
- Spread and overlap — Two groups with identical means can have completely different variances.
- Individual observations — Outliers, clusters, and gaps vanish behind a single bar.
- Sample size — A mean from 5 observations looks the same as a mean from 5,000.
⚠️ The Bar Chart Trap
When you see group averages in a dashboard, ask: "What does the distribution look like?" Use this studio's comparison mode to see how much information a bar chart of means discards compared to a violin or strip plot of the same data.
📊 Chart Type Trade-Offs
| Chart Type | Reveals | Hides | Best When |
|---|---|---|---|
| Bar Chart | Counts, averages, totals | Distribution shape, individual values | Audience needs a simple comparison |
| Scatter Plot | Relationship direction, clusters, outliers | Density in crowded regions (overplotting) | Exploring continuous × continuous |
| Box Plot | Median, quartiles, outliers, spread | Distribution shape (bimodality) | Quick group comparison with robust stats |
| Violin Plot | Full distribution shape, density peaks | Individual observations, exact values | Groups may have different shapes, not just different centers |
| Strip / Jitter Plot | Every individual data point, sample size | Density (when n is large, points overlap) | Small-to-medium datasets where each observation matters |
| Heatmap | Patterns across two dimensions, magnitude via color | Exact values (color is imprecise), distribution | Many categories or binned continuous data |
| Stacked Bar | Part-to-whole composition, proportions | Absolute counts (in 100% mode), individual category comparison | Showing how proportions shift across groups |