library(tidyverse)
library(scales)
library(ggthemes)
library(knitr)
library(kableExtra)
library(gridExtra)
library(corrplot)
library(ggrepel)

df <- read_csv("cleaned_social_media_advertising.csv") %>%
  mutate(
    acquisition_cost = as.numeric(gsub("\\$|,", "", acquisition_cost)),
    date             = as.Date(date),
    month            = floor_date(date, "month")
  )

1 Executive Summary

1.1 Dataset Overview

tibble(
  Metric  = c("Total campaigns", "Channels analyzed",
               "Customer segments", "Cities covered", "Time period"),
  Value   = c(
    format(nrow(df), big.mark = ","),
    n_distinct(df$channel_used),
    n_distinct(df$customer_segment),
    n_distinct(df$location),
    paste(min(df$date), "→", max(df$date))
  )
) %>%
  kable(align = c("l", "r"), caption = "Dataset dimensions") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE)

Dataset dimensions
Metric	Value
Total campaigns	300,000
Channels analyzed	4
Customer segments	5
Cities covered	5
Time period	2022-01-01 → 2022-12-31

1.2 Global KPIs

tibble(
  KPI   = c("Avg. Conversion Rate", "Avg. ROI", "Avg. Clicks",
             "Avg. Impressions", "Avg. Engagement Score", "Avg. Acquisition Cost"),
  Value = c(
    percent(mean(df$conversion_rate), accuracy = 0.1),
    as.character(round(mean(df$roi), 2)),
    format(round(mean(df$clicks), 0), big.mark = ","),
    format(round(mean(df$impressions), 0), big.mark = ","),
    as.character(round(mean(df$engagement_score), 2)),
    dollar(mean(df$acquisition_cost, na.rm = TRUE))
  )
) %>%
  kable(align = c("l", "r"), caption = "Global KPIs across all campaigns") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"),
                full_width = FALSE)

Global KPIs across all campaigns
KPI	Value
Avg. Conversion Rate	8.0%
Avg. ROI	3.18
Avg. Clicks	18,154
Avg. Impressions	56,034
Avg. Engagement Score	4.37
Avg. Acquisition Cost	$7,753.75

2 Channel Performance

2.1 Conversion Rate by Channel

df %>%
  group_by(channel_used) %>%
  summarise(
    conversion_mean = mean(conversion_rate),
    roi_mean        = mean(roi),
    n               = n(),
    .groups = "drop"
  ) %>%
  ggplot(aes(x = reorder(channel_used, -conversion_mean),
             y = conversion_mean, fill = channel_used)) +
  geom_col(width = 0.6, show.legend = FALSE) +
  geom_text(aes(label = percent(conversion_mean, accuracy = 0.1)),
            vjust = -0.5, size = 4, fontface = "bold") +
  scale_y_continuous(labels = percent_format(),
                     expand = expansion(mult = c(0, .12))) +
  scale_fill_brewer(palette = "Set2") +
  labs(title    = "Average Conversion Rate by Channel",
       subtitle = "Which platform converts best?",
       x = NULL, y = "Conversion Rate") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

2.2 ROI Distribution by Channel

df %>%
  ggplot(aes(x = channel_used, y = roi, fill = channel_used)) +
  geom_violin(alpha = 0.6, show.legend = FALSE) +
  geom_boxplot(width = 0.15, fill = "white", outlier.shape = NA) +
  scale_fill_brewer(palette = "Set2") +
  labs(title    = "ROI Distribution by Channel",
       subtitle = "Violin + boxplot to show real spread",
       x = NULL, y = "ROI") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

2.3 Engagement Score by Channel

df %>%
  group_by(channel_used) %>%
  summarise(eng_mean = mean(engagement_score), .groups = "drop") %>%
  ggplot(aes(x = reorder(channel_used, eng_mean), y = eng_mean, fill = channel_used)) +
  geom_col(show.legend = FALSE) +
  geom_text(aes(label = round(eng_mean, 2)), hjust = -0.2, fontface = "bold") +
  coord_flip() +
  scale_fill_brewer(palette = "Pastel1") +
  scale_y_continuous(expand = expansion(mult = c(0, .15))) +
  labs(title = "Average Engagement Score by Channel",
       x = NULL, y = "Engagement Score") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

3 Campaign Goal vs. Results

3.1 Conversion & ROI by Campaign Goal

df %>%
  group_by(campaign_goal) %>%
  summarise(
    conv    = mean(conversion_rate),
    roi     = mean(roi),
    .groups = "drop"
  ) %>%
  pivot_longer(c(conv, roi), names_to = "metric", values_to = "value") %>%
  mutate(metric = recode(metric, conv = "Conversion Rate", roi = "ROI")) %>%
  ggplot(aes(x = reorder(campaign_goal, -value), y = value, fill = campaign_goal)) +
  geom_col(show.legend = FALSE) +
  facet_wrap(~metric, scales = "free_y") +
  scale_fill_brewer(palette = "Set3") +
  labs(title = "Conversion Rate & ROI by Campaign Goal",
       x = NULL, y = NULL) +
  theme_minimal(base_size = 12) +
  theme(
    plot.title  = element_text(face = "bold"),
    axis.text.x = element_text(angle = 20, hjust = 1)
  )

3.2 Channel × Goal — ROI Heatmap

df %>%
  group_by(channel_used, campaign_goal) %>%
  summarise(roi_mean = mean(roi), .groups = "drop") %>%
  ggplot(aes(x = channel_used, y = campaign_goal, fill = roi_mean)) +
  geom_tile(color = "white", linewidth = 0.8) +
  geom_text(aes(label = round(roi_mean, 2)), size = 4, fontface = "bold") +
  scale_fill_gradient(low = "#fee8c8", high = "#d7301f", name = "Avg.\nROI") +
  labs(title    = "Heatmap: Average ROI — Channel × Campaign Goal",
       subtitle = "Darker cells = higher return",
       x = "Channel", y = "Campaign Goal") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

4 Customer Segment Analysis

4.1 ROI by Segment

df %>%
  ggplot(aes(x = reorder(customer_segment, roi, FUN = median),
             y = roi, fill = customer_segment)) +
  geom_boxplot(show.legend = FALSE, outlier.alpha = 0.1) +
  coord_flip() +
  scale_fill_brewer(palette = "Set2") +
  labs(title    = "ROI Distribution by Customer Segment",
       subtitle = "Which segment generates the most return?",
       x = NULL, y = "ROI") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

4.2 Conversion Rate by Segment & Channel

df %>%
  group_by(customer_segment, channel_used) %>%
  summarise(conv = mean(conversion_rate), .groups = "drop") %>%
  ggplot(aes(x = channel_used, y = conv, color = customer_segment,
             group = customer_segment)) +
  geom_line(linewidth = 1.2) +
  geom_point(size = 3) +
  scale_y_continuous(labels = percent_format()) +
  scale_color_brewer(palette = "Dark2", name = "Segment") +
  labs(title    = "Conversion Rate: Channel × Customer Segment",
       subtitle = "How conversion varies per channel across segments",
       x = "Channel", y = "Conversion Rate") +
  theme_minimal(base_size = 13) +
  theme(plot.title      = element_text(face = "bold"),
        legend.position = "bottom")

5 Target Audience

5.1 Conversion vs. ROI by Audience

df %>%
  group_by(target_audience) %>%
  summarise(
    conv = mean(conversion_rate),
    roi  = mean(roi),
    .groups = "drop"
  ) %>%
  ggplot(aes(x = conv, y = roi, label = target_audience, color = target_audience)) +
  geom_point(size = 5, show.legend = FALSE) +
  geom_text_repel(size = 3.5, show.legend = FALSE, max.overlaps = 20) +
  scale_x_continuous(labels = percent_format()) +
  scale_color_brewer(palette = "Paired") +
  labs(title    = "Conversion Rate vs. ROI by Target Audience",
       subtitle = "Top-right quadrant = best overall performance",
       x = "Conversion Rate", y = "Average ROI") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

6 Geography & Language

6.1 ROI by City

df %>%
  group_by(location) %>%
  summarise(roi_mean = mean(roi), .groups = "drop") %>%
  ggplot(aes(x = reorder(location, roi_mean), y = roi_mean, fill = location)) +
  geom_col(show.legend = FALSE) +
  geom_text(aes(label = round(roi_mean, 2)), hjust = -0.2, fontface = "bold") +
  coord_flip() +
  scale_fill_brewer(palette = "Blues") +
  scale_y_continuous(expand = expansion(mult = c(0, .15))) +
  labs(title = "Average ROI by City",
       x = NULL, y = "Average ROI") +
  theme_minimal(base_size = 13) +
  theme(plot.title = element_text(face = "bold"))

6.2 Preferred Channel by Language

df %>%
  count(language, channel_used) %>%
  group_by(language) %>%
  mutate(prop = n / sum(n)) %>%
  ungroup() %>%
  ggplot(aes(x = language, y = prop, fill = channel_used)) +
  geom_col(position = "fill") +
  scale_y_continuous(labels = percent_format()) +
  scale_fill_brewer(palette = "Set2", name = "Channel") +
  labs(title    = "Channel Distribution by Campaign Language",
       subtitle = "Which channel dominates in each language?",
       x = "Language", y = "Proportion") +
  theme_minimal(base_size = 13) +
  theme(plot.title      = element_text(face = "bold"),
        legend.position = "bottom")

7 Time Trends

7.1 Monthly ROI by Channel

df %>%
  group_by(month, channel_used) %>%
  summarise(roi_mean = mean(roi), .groups = "drop") %>%
  ggplot(aes(x = month, y = roi_mean, color = channel_used, group = channel_used)) +
  geom_line(linewidth = 1) +
  geom_smooth(se = FALSE, linetype = "dashed", linewidth = 0.5) +
  scale_color_brewer(palette = "Set1", name = "Channel") +
  scale_x_date(date_labels = "%b %Y", date_breaks = "3 months") +
  labs(title    = "Monthly ROI Trend by Channel",
       subtitle = "Dashed line = smoothed trend",
       x = NULL, y = "Average ROI") +
  theme_minimal(base_size = 13) +
  theme(
    plot.title      = element_text(face = "bold"),
    axis.text.x     = element_text(angle = 30, hjust = 1),
    legend.position = "bottom"
  )

7.2 Monthly Campaign Volume

df %>%
  count(month) %>%
  ggplot(aes(x = month, y = n)) +
  geom_area(fill = "#3182bd", alpha = 0.4) +
  geom_line(color = "#3182bd", linewidth = 1) +
  scale_x_date(date_labels = "%b %Y", date_breaks = "3 months") +
  scale_y_continuous(labels = comma) +
  labs(title = "Monthly Campaign Volume",
       x = NULL, y = "Number of Campaigns") +
  theme_minimal(base_size = 13) +
  theme(
    plot.title  = element_text(face = "bold"),
    axis.text.x = element_text(angle = 30, hjust = 1)
  )

8 Correlations & Numeric Metrics

8.1 Correlation Matrix

num_vars <- df %>%
  select(conversion_rate, roi, clicks, impressions,
         engagement_score, acquisition_cost) %>%
  drop_na()

M <- cor(num_vars)

corrplot(M,
         method      = "color",
         type        = "upper",
         addCoef.col = "black",
         tl.col      = "black",
         tl.srt      = 45,
         col         = colorRampPalette(c("#d73027", "#f7f7f7", "#4575b4"))(200),
         title       = "Correlation Matrix — Numeric Variables",
         mar         = c(0, 0, 2, 0))

8.2 Clicks vs. Conversion Rate (by Channel)

df %>%
  sample_n(5000) %>%
  ggplot(aes(x = clicks, y = conversion_rate, color = channel_used)) +
  geom_point(alpha = 0.4, size = 1.5) +
  geom_smooth(method = "lm", se = FALSE, linewidth = 1.2) +
  scale_y_continuous(labels = percent_format()) +
  scale_x_continuous(labels = comma) +
  scale_color_brewer(palette = "Set1", name = "Channel") +
  labs(title    = "Clicks vs. Conversion Rate by Channel",
       subtitle = "Random sample of 5,000 campaigns",
       x = "Clicks", y = "Conversion Rate") +
  theme_minimal(base_size = 13) +
  theme(plot.title      = element_text(face = "bold"),
        legend.position = "bottom")

9 Comparative Summary Table

df %>%
  group_by(channel_used, campaign_goal) %>%
  summarise(
    Campaigns      = n(),
    `Conv. Rate`   = percent(mean(conversion_rate), accuracy = 0.1),
    `Avg. ROI`     = round(mean(roi), 2),
    `Eng. Score`   = round(mean(engagement_score), 1),
    .groups = "drop"
  ) %>%
  rename(Channel = channel_used, Goal = campaign_goal) %>%
  arrange(Channel, desc(`Avg. ROI`)) %>%
  kable(align   = c("l", "l", "r", "r", "r", "r"),
        caption = "Summary Table: Channel × Campaign Goal") %>%
  kable_styling(
    bootstrap_options = c("striped", "hover", "condensed", "responsive"),
    full_width = TRUE
  ) %>%
  row_spec(0, bold = TRUE, background = "#2c3e50", color = "white")

Summary Table: Channel × Campaign Goal
Channel	Goal	Campaigns	Conv. Rate	Avg. ROI	Eng. Score
Facebook	Product Launch	18775	8.0%	4.00	5.5
Facebook	Brand Awareness	18836	8.0%	3.99	5.5
Facebook	Increase Sales	18663	8.0%	3.98	5.5
Facebook	Market Expansion	18890	8.0%	3.98	5.5
Instagram	Product Launch	18874	7.9%	4.02	5.5
Instagram	Brand Awareness	18774	8.0%	4.01	5.5
Instagram	Increase Sales	18690	7.9%	4.01	5.5
Instagram	Market Expansion	18763	8.0%	4.00	5.5
Pinterest	Brand Awareness	18918	8.0%	0.72	1.0
Pinterest	Increase Sales	18732	8.0%	0.72	1.0
Pinterest	Product Launch	18814	8.0%	0.72	1.0
Pinterest	Market Expansion	18618	8.0%	0.71	1.0
Twitter	Product Launch	18567	8.0%	4.04	5.5
Twitter	Increase Sales	18878	8.0%	4.01	5.5
Twitter	Brand Awareness	18720	8.0%	3.99	5.5
Twitter	Market Expansion	18488	8.1%	3.97	5.5

10 Key Conclusions

Channel with highest average ROI: Instagram
Most profitable customer segment: Technology
Campaign goal with highest conversion: Market Expansion

Key takeaways for your presentation:

All four channels show very similar ROI distributions, suggesting that content strategy and audience targeting matter more than channel selection alone.
Engagement score has no strong linear correlation with ROI, indicating that surface-level interaction does not guarantee financial return.
Product Launch and Increase Sales campaigns consistently achieve higher conversion rates, as the goal aligns directly with the expected user action.
Geographic variation is moderate, but Los Angeles and New York show the highest campaign volume and the most stable performance.
Conversion rate varies primarily by target audience rather than language — challenging common assumptions about localization strategy.

Analysis automatically generated in R · 2026-04-19

Social Media Advertising Analytics

Comparative Analysis of Digital Campaigns

Marketing Analytics

2026-04-19

1 Executive Summary

1.1 Dataset Overview

1.2 Global KPIs

2 Channel Performance

2.1 Conversion Rate by Channel

2.2 ROI Distribution by Channel

2.3 Engagement Score by Channel

3 Campaign Goal vs. Results

3.1 Conversion & ROI by Campaign Goal

3.2 Channel × Goal — ROI Heatmap

4 Customer Segment Analysis

4.1 ROI by Segment

4.2 Conversion Rate by Segment & Channel

5 Target Audience

5.1 Conversion vs. ROI by Audience

6 Geography & Language

6.1 ROI by City

6.2 Preferred Channel by Language

7 Time Trends

7.1 Monthly ROI by Channel

7.2 Monthly Campaign Volume

8 Correlations & Numeric Metrics

8.1 Correlation Matrix

8.2 Clicks vs. Conversion Rate (by Channel)

9 Comparative Summary Table

10 Key Conclusions