Skip to main content

Chart Types

A complete gallery of visualizations you can create—from simple bars to complex Sankey diagrams.

Overview​

Shadowfax uses the Vega visualization grammar, giving you access to virtually unlimited chart types. Just describe what you want in natural language, and the AI generates it. No need to know Vega syntax or charting libraries.

Chart Types Gallery

Sample gallery of available chart types

Basic Charts​

Bar Charts​

What they're for: Comparing categories or groups

Variations:

  • Vertical bars (column charts)
  • Horizontal bars
  • Grouped bars (multiple series side-by-side)
  • Stacked bars (segments within bars)

Example request:

@[sales_by_region] Create a horizontal bar chart sorted by revenue

Bar Chart

Horizontal bar chart comparing regions

Line Charts​

What they're for: Trends over time or continuous data

Variations:

  • Single line
  • Multiple lines (comparing series)
  • With markers
  • Smoothed curves

Example request:

@[monthly_revenue] Show as a line chart with markers for each month

Line Chart

Line chart showing trend over time

Area Charts​

What they're for: Showing magnitude over time with filled regions

Variations:

  • Single area
  • Stacked areas (cumulative)
  • Overlapping areas (with transparency)

Example request:

@[product_revenue] Create a stacked area chart showing each product's contribution over time

Area Chart

Stacked area chart showing composition over time

Scatter Plots​

What they're for: Showing relationships between two variables

Variations:

  • Simple scatter
  • With size encoding (bubble chart)
  • With color encoding
  • With trend lines

Example request:

@[customers] Create a scatter plot of customer age vs. lifetime value

Scatter Plot

Scatter plot showing correlation between variables

Pie and Donut Charts​

What they're for: Part-to-whole relationships

When to use: Small number of categories (3-6 ideal)

Example request:

@[market_share] Show as a donut chart with percentages labeled

Donut Chart

Donut chart showing proportions

Statistical Charts​

Histograms​

What they're for: Distribution of a single variable

Example request:

@[order_values] Create a histogram showing the distribution of order amounts

Histogram

Histogram showing value distribution

Box Plots​

What they're for: Showing median, quartiles, and outliers

Example request:

@[response_times] Create box plots for each service endpoint

Box Plot

Box plots showing distribution statistics

Violin Plots​

What they're for: Distribution shape and density

Example request:

@[purchase_amounts] Show violin plots by customer segment

Advanced Visualizations​

Heatmaps​

What they're for: Two-dimensional density or correlation matrices

Example request:

@[hourly_traffic] Create a heatmap with days of week on Y-axis and hours on X-axis

Heatmap

Heatmap showing patterns across two dimensions

Treemaps​

What they're for: Hierarchical data with size-based rectangles

Example request:

@[revenue_by_category] Show as a treemap sized by revenue, colored by profit margin

Treemap

Treemap showing hierarchical proportions

Sunburst Diagrams​

What they're for: Multi-level hierarchical data in concentric circles

Example request:

@[category_hierarchy] Create a sunburst diagram showing Category > Subcategory > Product

Sunburst

Sunburst diagram showing nested hierarchies

Sankey Diagrams​

What they're for: Flow and transitions between stages

Common uses: Customer journeys, process flows, budget allocation

Example request:

@[user_journey] Create a Sankey diagram showing flow from Landing Page → Category → Purchase

Sankey

Sankey diagram showing flows between stages

Network Diagrams​

What they're for: Relationships and connections between entities

Example request:

@[user_interactions] Show as a network diagram with nodes for users and edges for connections

Network

Network graph showing connections

Geographic Visualizations​

Choropleth Maps​

What they're for: Data values across geographic regions

Example request:

@[sales_by_state] Create a US map colored by revenue

Choropleth Map

Map with regions colored by data values

Bubble Maps​

What they're for: Point data with size encoding

Example request:

@[store_locations] Show stores on a map with bubble size representing revenue

Specialized Charts​

Waterfall Charts​

What they're for: Showing cumulative effect of sequential values

Common uses: Financial changes, variance analysis

Example request:

@[budget_changes] Create a waterfall chart showing monthly additions and subtractions

Waterfall

Waterfall chart showing sequential changes

Bullet Charts​

What they're for: Progress toward a goal with context

Example request:

@[kpis] Show each KPI as a bullet chart with target and actual

Gauge Charts​

What they're for: Single metric with min/max range

Example request:

@[customer_satisfaction] Show as a gauge from 0-100

Combination Charts​

Mix multiple chart types:

Example request:

@[monthly_data] Show revenue as bars and profit margin as a line on the same chart

Combo Chart

Bar and line chart combined

How to Request Chart Types​

Be Specific​

Good: "Create a horizontal bar chart sorted by value descending" Less specific: "Make a chart"

Describe the Visual​

Good: "Show as a stacked area chart with each product as a colored layer" Less specific: "Visualize products over time"

Reference Examples​

Good: "Create a Sankey diagram like a customer journey flow" Good: "Show as a treemap similar to a disk space visualizer"

Choosing the Right Chart​

Comparing categories → Bar chart

Trends over time → Line or area chart

Part-to-whole → Pie, donut, or treemap

Distribution → Histogram or box plot

Relationship between variables → Scatter plot

Geographic patterns → Choropleth or bubble map

Flow between stages → Sankey diagram

Hierarchy → Treemap or sunburst

Connections → Network diagram

Tips & Best Practices​

Start with standard types: Bar and line charts work for 80% of use cases.

Match chart to data structure: Table with categories → bar chart. Time series → line chart.

Consider your audience: Stick to familiar chart types for general audiences.

Experiment freely: Requesting a chart type is fast—try multiple options.

Ask for alternatives: "What chart types would work best for this data?"

Specify orientation: "Horizontal bar chart" vs. "vertical bar chart"

Combine when useful: Don't be afraid to layer bars and lines or combine visualizations.