Module 5 Lab

Coupons Used per Income Bracket

#Libraries
library(completejourney)
library(dplyr)
library(ggplot2)

#Data
transactions <- get_transactions()
promotions <- get_promotions()
demographics <- demographics
coupon_redemptions <-coupon_redemptions
campaigns <- campaigns

# Data Cleaning
demographics$income <- gsub("[^0-9.]", "", demographics$income)
demographics$income <- as.numeric(demographics$income)

#Remove NA
demographics <- demographics[!is.na(demographics$income), ]

# Income Brackets
demographics$income_bracket <- cut(demographics$income,
                                   breaks = c(0, 50000, 100000, 150000, 200000, 250000, Inf),
                                   labels = c("> $50K", "$50K-$100K", "$100K-$150K", "$150K-$200K", "$200K-$250K", ">$250K"),
                                   right = FALSE)
# Merge
household_coupons <- coupon_redemptions %>%
  inner_join(demographics, by = "household_id") %>%
  group_by(income_bracket) %>%
  summarize(total_coupons_used = n())

# Visualization 
p <- ggplot(household_coupons, aes(x = income_bracket, y = total_coupons_used, fill = income_bracket)) +
  geom_bar(stat = "identity", fill = "grey80") +
  geom_point(aes(color = income_bracket), size = 3)  
p + geom_text(aes(label = total_coupons_used), vjust = -0.5) +
  annotate("text", x = 2, y = 1500, label = "Highest usage here", color = "black", size = 3, angle = 360, fontface = "italic")

Spending by Household Composition

household_spending <- transactions %>%
  inner_join(demographics, by = "household_id") %>%
  group_by(household_comp) %>%
  summarize(average_spending = mean(sales_value, na.rm = TRUE))

#Order
household_spending$household_comp <- factor(household_spending$household_comp, 
                                            levels = unique(demographics$household_comp[order(demographics$household_size)]))


#Visualization
spending_plot <- ggplot(household_spending, aes(x = household_comp, y = average_spending)) +
  geom_col(fill = "grey80", color = "black") +  
  geom_point(aes(y = average_spending), color = "red", size = 3, show.legend = FALSE) +  
  geom_text(aes(label = round(average_spending, 2)), vjust = -2, color = "black") +  
  labs(title = "Average Household Spending by Composition",
       x = "Household Composition",
       y = "Average Spending ($)") +
  theme_minimal() +
  theme(
    plot.title = element_text(face = "bold", hjust = 0.5, size = 16),
    axis.title = element_text(face = "bold", size = 12),
    axis.text.x = element_text(angle = 45, hjust = 1, vjust = 1),
    axis.text.y = element_text(color = "black"),
    panel.background = element_rect(fill = "white", colour = "grey50"),  # 
    panel.grid.major = element_line(color = "grey90"),  
    panel.grid.minor = element_blank()  
  )
print(spending_plot)

Coupon Usage Between Households With and Without Children

#Filter
demographics$has_children <- ifelse(demographics$kids_count > 0, "With Children", "No Children")

# Merge
household_coupons <- coupon_redemptions %>%
  inner_join(demographics, by = "household_id") %>%
  group_by(has_children) %>%
  summarize(total_coupons_used = n(), num_households = n_distinct(household_id), .groups = 'drop') %>%
  mutate(average_coupons_per_household = total_coupons_used / num_households)
#Visualization
colors_for_children_status <- c("With Children" = "#56B4E9", "No Children" = "#E69F00")

coupon_usage_plot <- ggplot(household_coupons, aes(x = has_children, y = average_coupons_per_household, fill = has_children)) +
  geom_col(show.legend = FALSE) +  
  scale_fill_manual(values = colors_for_children_status) + 
  geom_text(aes(label = round(average_coupons_per_household, 2)), vjust = -0.5, color = "white") + 
  labs(title = "Average Coupon Usage per Household by Child Status",
       x = "Household Child Status",
       y = "Average Coupons Used per Household") +
  theme_minimal() +
  theme(
    plot.title = element_text(face = "bold", hjust = 0.5, size = 16),
    axis.title = element_text(face = "bold", size = 12),
    axis.text.x = element_text(angle = 0, hjust = 0.5),
    axis.text.y = element_text(color = "black"),
    panel.background = element_rect(fill = "white", colour = "grey50"),  
    panel.grid.major = element_line(color = "grey90"),  
    panel.grid.minor = element_blank()  
  )

print(coupon_usage_plot)