End of course project for Google Data Analytics Professional Certificate

Date Created: Jul 29, 2023

Goal: find the reasons behind users abandoning carts.

Tools used: Tableau, Power BI, and R language.

This is fictitious data created by me that could represent a real-world situation.

I approached this project by following the data analytics phases outlined in the certificate program:

1. Ask

• What are the possible reasons behind abandoning carts?
• How can we reduce them?

2. Prepare, 3. Process

The data is fictitious data created by me and it could represent a real-world case, it is very small data that consists of 19 rows and 8 columns, here is a sample of the data:

4. Analyze

I analyzed the data to find possible reasons, here are some pictures of the process using Tableau:

We can see the average cart value for those who completed order is significantly higher than who did not complete the order:

Here we can see that 75% of users who did not complete the order have added their payment info, which means there is no problem in the payment page:

Here we can see that 83.33% of users who did not complete the order have NOT added their address info, which makes me think about a possible problem in the shipping prices, as the shipping rates only show up AFTER the payment info is added:

Here is the last chart to show, which shows that most of the completed orders are above $40 in cart value, which makes my last point true:

Wrap-Up:

• The customers are getting shocked by the shipping prices, that's why most of customers whose cart's value is less than $40 are abandoning their carts, because the shipping itself could be +30% of their cart value, so they find it not worth it to buy.

5. Share, 6. Act

We should investigate our shipping rates to solve the problem, my suggestion is to make the shipping free or $5 max and slightly increase the prices of all products like 5% for example to cover the shipping costs, this needs further studying by sales team to determine how much exactly the products prices should be increased.

Done, I also worked on this project using R language too, here is the code and the output:

library(readr)
library(dplyr)
library(skimr)
library(janitor)
library(ggplot2)
library(tidyverse)
library(here)
install.packages("egg")
library(egg)

abandoned_carts <- read.csv("C:/Users/s2ahm/Desktop/Data Analytics/Google Data Analytics Professional Certificate/Case Study/Abandoned carts.csv")

view(abandoned_carts)
str(abandoned_carts)
count_completed_order <- table(abandoned_carts$completed_order)
count_completed_order

y = prop.table(stat(count))
label = scales::percent(prop.table(after_stat(count)))

value_to_count <- 19
count <- length(which(abandoned_carts$cart_value == value_to_count))
completed_order_sum <- sum(abandoned_carts$completed_order)
completed_order_count <- summary(abandoned_carts$completed_order)
mutate(abandoned_carts, completed_order_sum)
completed_order_count
sum(abandoned_carts$completed_order, na.rm = TRUE)
count_user <- table(abandoned_carts$user)
count_user2 <- summary(abandoned_carts$user)
count_user2

plot1 <- 
  ggplot(data = abandoned_carts, 
  mapping = aes(x = (group = cut(cart_value, breaks = seq(0, 120, 20))), y = prop.table(stat(count)), 
  fill = completed_order)) +
  geom_col() +
  geom_text(aes(label=count_user), vjust=3) +
  labs(title = "First R Plot", caption = "abandoned carts", x = "Cart Value", y = "Number of Users")
  
seq_cart_value <- abandoned_carts %>%
  group_by(group = cut(cart_value, breaks = seq(0, 120, 20))) %>%
  summarize(n = n())
       
scale_x_continuous("",breaks=c(2.5,6,9),labels=c("Group1","Group2","Group3"))

ann_text <- data.frame(mpg = 15,wt = 5,lab = "Text",
                       cyl = factor(8,levels = c("4","6","8")))
p + geom_text(data = ann_text,label = "Text")

p <- ggplot(abandoned_carts, aes(x = (group = cut(cart_value, breaks = seq(0, 120, 20),

  labels=c("$0-19", "$20-39", "$40-59", "$60-79", "$80-99", "$100-119"))), fill = completed_order)) +

  geom_bar(stat = "count") +

  scale_fill_manual(name="Abandoned Carts",

                    labels=c("Abandoned Cart", "Completed Order"),

                    values=c("#990012", "#77BFC7")) +

  labs(title = "Abandoned Carts", x = "Cart Value", y = "Count of Users") +

  theme(plot.title = element_text(hjust = 0.5)) +

  geom_text(aes(label=after_stat(count)), stat="count", vjust=1.5) +

  #annotate("text", x = 5, y = 6, label = "order completion rate 

           #for\ncart_value<40 is 16.66%\norder completion rate for cart_value>40 is 71.43%", color = "", size = 0.8) +

  geom_vline(xintercept = 2.5, linetype="solid", color = "navy") +

  facet_wrap(~completed_order, nrow=2)

my_tag = c("order completion rate for\ncart_value<40 is 16.66%", "order completion rate for\ncart_value>40 is 71.43%")

tag_facet(p, 

          x = -Inf, y = -Inf, 

          vjust = -4, hjust = -1.25,

          open = "", close = "",

          fontface = 1,

          size = 4,

          tag_pool = my_tag)

output:

and here is a Power BI sheet I made too:

Done.

Thanks for reading!