Nested Column Bar Chart

Overview

This code creates an example Nested Column Chart of the type developed by Brittany Rosenau.

Setup

library(ggplot2)
library(dplyr)
library(readr)

Create Sample Data

csv_data <- "Region,Segment,Revenue
West,Consumer,364
West,Corporate,232
West,Home Office,143
East,Consumer,357
East,Corporate,204
East,Home Office,131
Central,Consumer,254
Central,Corporate,158
Central,Home Office,91
South,Consumer,196
South,Corporate,122
South,Home Office,74"

df <- read_csv(csv_data, show_col_types = FALSE)
df

# A tibble: 12 × 3
   Region  Segment     Revenue
   <chr>   <chr>         <dbl>
 1 West    Consumer        364
 2 West    Corporate       232
 3 West    Home Office     143
 4 East    Consumer        357
 5 East    Corporate       204
 6 East    Home Office     131
 7 Central Consumer        254
 8 Central Corporate       158
 9 Central Home Office      91
10 South   Consumer        196
11 South   Corporate       122
12 South   Home Office      74

Data Preparation

# Ensure Region is a factor with the desired order
df$Region <- factor(df$Region, levels = c("West", "East", "Central", "South"))

# Ensure Segment is a factor with the desired order for legend/coloring
df$Segment <- factor(df$Segment, levels = c("Consumer", "Corporate", "Home Office"))

# Calculate total revenue per region
region_totals <- df %>%
    group_by(Region) %>%
    summarise(TotalRevenue = sum(Revenue), .groups = "drop")

region_totals

# A tibble: 4 × 2
  Region  TotalRevenue
  <fct>          <dbl>
1 West             739
2 East             692
3 Central          503
4 South            392

Define Custom Colors

segment_colors <- c(
    "Consumer" = "#2ECC71",    # Green
    "Corporate" = "#F39C12",   # Orange
    "Home Office" = "#3498DB"  # Blue
)

Create the Nested Column Chart

p <- ggplot() +
    # Layer 1: Grey background bars representing total revenue per region
    geom_col(
        data = region_totals,
        aes(x = Region, y = TotalRevenue),
        fill = "grey85",
        alpha = 0.8,
        width = 0.8
    ) +
    # Layer 2: Dodged bars for segment revenue
    geom_col(
        data = df,
        aes(x = Region, y = Revenue, fill = Segment),
        position = position_dodge(width = 0.8),
        width = 0.7
    ) +
    # Layer 3: Labels for individual segment bars
    geom_text(
        data = df,
        aes(x = Region, y = Revenue, label = paste0("$", Revenue, "K"), group = Segment),
        position = position_dodge(width = 0.8),
        vjust = -0.5,
        size = 2.75,
        family = "sans"
    ) +
    # Layer 4: Labels for total region revenue
    geom_text(
        data = region_totals,
        aes(x = Region, y = TotalRevenue, label = paste0("$", TotalRevenue, "K")),
        vjust = -2.0,
        size = 3.5,
        family = "sans"
    ) +
    # Layer 5: Labels for Region Names
    geom_text(
        data = region_totals,
        aes(x = Region, y = TotalRevenue, label = as.character(Region)),
        vjust = -3.5,
        size = 4,
        fontface = "bold",
        family = "sans"
    ) +
    # Apply custom color scale
    scale_fill_manual(values = segment_colors) +
    # Set plot title
    ggtitle("Total Sales by Region and Segment ($K)") +
    # Customize theme
    theme_void(base_size = 12, base_family = "sans") +
    theme(
        plot.title = element_text(
            hjust = 0,
            face = "bold",
            size = 16,
            margin = margin(b = 15)
        ),
        legend.position = "top",
        legend.justification = "left",
        legend.title = element_blank(),
        legend.text = element_text(size = 11),
        legend.key.size = unit(0.5, "cm"),
        legend.margin = margin(b = -10),
        axis.text = element_blank(),
        axis.ticks = element_blank(),
        axis.title = element_blank(),
        plot.margin = margin(20, 20, 20, 20)
    ) +
    # Adjust y-axis limits for label space
    scale_y_continuous(
        limits = c(0, max(region_totals$TotalRevenue) * 1.35),
        expand = expansion(mult = c(0, 0))
    )

print(p)