###library(dplyr)
###library(ggplot2)
###library(readxl)
###library(tidyverse)
###library(plotly)
###library(viridis)
###library(colorspace)
###library(ggthemes)Big Idea
We plan to earn customer loyalty and increase revenue through machine-readable marketing data, which allows us to create a user behavior profile and more accurately predict customer behavior to use in forecasting revenue.
The Data
This data came from: https://www.kaggle.com/imakash3011/customer-personality-analysis
marketing_campaign_original <-read_excel("Mark_Campaign_Clean.xlsx")Convert data to tibble
marketing_campaign <- as_tibble(marketing_campaign_original)Dropping levels
## Remove Absurd, Alone and YOLO, Minor, Silent
marketing_campaign <- marketing_campaign %>%
filter(Marital_Status != "Absurd", Marital_Status != "Alone", Marital_Status != "YOLO", Age_Grp != "Minor", Age_Grp != "Silent")%>%
droplevels()Income by Age Group
gpl<- marketing_campaign %>%
filter(Income <= 150000)%>%
ggplot(aes(Age, Income), group_by= Age, colour= Age)+
geom_col()+
scale_y_continuous(labels = scales::comma)+
theme_economist()+
theme(
panel.grid.major = element_blank(),
panel.grid.minor.x = element_blank(),
legend.position = "bottom"
)
ggplotly(gpl)What products do people buy the most by generation?
p10 <- marketing_campaign %>%
select(c(Age_Grp, "MntWines", "MntFruits", "MntMeatProducts", "MntFishProducts", "MntSweetProducts", "MntGoldProds")) %>%
mutate(Wines = as.integer(MntWines),
Fruits= as.integer(MntFruits),
Meat = as.integer(MntMeatProducts),
Fish = as.integer(MntFishProducts),
Sweet = as.integer(MntSweetProducts),
Gold = as.integer(MntGoldProds))%>%
select(c(Age_Grp, Wines, Fruits, Meat, Fish, Sweet, Gold))%>%
pivot_longer(-(Age_Grp), names_to= "Products", values_to= "Value")%>%
ggplot(aes(x=Age_Grp, y=Value, fill=Age_Grp))+
geom_boxplot()+
theme(
axis.text.x = element_text(hjust = 0.8)
) +
facet_wrap(.~Products, scales = "free_y")
p10
What channels do people use the most to buy items?
marketing_campaign %>%
select(c(Age_Grp,"NumWebPurchases", "NumCatalogPurchases", "NumStorePurchases"))%>%
mutate(Online = as.integer(NumWebPurchases),
Catalog = as.integer(NumCatalogPurchases),
Store = as.integer(NumStorePurchases))%>%
select(c(Age_Grp, Online, Catalog, Store))%>%
pivot_longer(-(Age_Grp), names_to = "Channels", values_to= "Value") %>%
filter(Value < 20)%>%
ggplot(aes(x=Channels, y=Value, fill = Channels))+
geom_violin(trim=FALSE)+
geom_boxplot(width=0.1, fill="white")+
facet_wrap(.~Age_Grp, scales = "free_y")+
theme_wsj()+
labs(y="")+
theme(
legend.position = "none")
What is the relationship between Income and Average Purchase per order and Education?
p12 <- marketing_campaign %>%
filter(Avg_Purchase_Per_Order < 100)%>%
filter(Income < 150000)%>%
mutate(Avg_Purchase_Per_Order=round(Avg_Purchase_Per_Order)) %>%
arrange((Kidhome)) %>%
ggplot(aes(Avg_Purchase_Per_Order, Income, fill= Education))+
geom_point(alpha=0.5, shape=21,
position = "jitter") +
geom_density_2d_filled(alpha = 0.5) +
geom_density_2d(size = 0.25, colour = "yellow")+
facet_grid(.~Education)+
theme(
axis.text.y = element_text(size = 8),
axis.text.x = element_text(size= 8),
plot.caption.position = "panel",
plot.caption = element_text(hjust = 0, size = 7),
strip.text.x = element_text(face = "bold", size= 10),
panel.background = element_blank(),
panel.border = element_blank(),
panel.spacing = unit(1, "lines"),
legend.title = element_blank(),
legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor.x = element_blank()
)
ggplotly(p12)Do people who have children at home spend more money in wine?
ggplot(marketing_campaign, aes(as.factor(Kidhome), MntWines, color= as.factor(Kidhome)))+
geom_jitter(aes(size= MntWines, alpha= 0.1))+
scale_alpha(guide = 'none')+
scale_size("Amount of\nMoney", range= c(1,5))+
scale_color_viridis_d(option = "viridis", name= "Number of\nKids")+
labs( x = "Number of Kids at Home",
y = "Amount Spent on Wine")+
theme(
axis.line.y=element_blank(),
axis.line.x = element_blank(),
axis.ticks = element_blank(),
axis.text.y = element_text(size = 6.5, ),
axis.text.x = element_text(size= 6.5),
panel.grid.minor.y = element_blank(),
panel.grid.major.y = element_line(size= 0.1, color= "white",
linetype= "solid"),
panel.background = element_rect(fill = "lightgrey"),
panel.border = element_blank(),
panel.spacing = unit(1, "lines"),
strip.background= element_rect(fill= "white", linetype = "blank"),
strip.text = element_text(color= "black", face= "bold"),
strip.text.x = element_text(face = "bold", size= 10),
legend.position = "bottom",
legend.margin=margin(grid::unit(0,"cm")),
legend.key.width=grid::unit(0.5,"cm"),
legend.key.height=grid::unit(0.5,"cm"))+
facet_grid(~Age_Grp)
What about people that have teenagers at home?
ggplot(marketing_campaign, aes(as.factor(Teenhome), MntWines, color= as.factor(Teenhome)))+
geom_jitter(aes(size= MntWines, alpha= 0.1))+
scale_alpha(guide = 'none')+
scale_size("Amount of\nMoney", range= c(1,5))+
scale_color_viridis_d(option = "viridis", name= "Number of\nTeens")+
labs( x = "Number of Teenagers at Home",
y = "Amount Spent on Wine")+
theme(
axis.line.y=element_blank(),
axis.line.x = element_blank(),
axis.ticks = element_blank(),
axis.text.y = element_text(size = 6.5, ),
axis.text.x = element_text(size= 6.5),
panel.grid.minor.y = element_blank(),
panel.grid.major.y = element_line(size= 0.1, color= "white",
linetype= "solid"),
panel.background = element_rect(fill = "lightgrey"),
panel.border = element_blank(),
panel.spacing = unit(1, "lines"),
strip.background= element_rect(fill= "white", linetype = "blank"),
strip.text = element_text(color= "black", face= "bold"),
strip.text.x = element_text(face = "bold", size= 10),
legend.position = "bottom",
legend.margin=margin(grid::unit(0,"cm")),
legend.key.width=grid::unit(0.5,"cm"),
legend.key.height=grid::unit(0.5,"cm")
)+
facet_grid(~Age_Grp)