Exploratory Data Analysis on Shopping Trends in Python

Data Analysis Project by Anushka Sharma, Team edSlash.

Customer Shopping Trends

Customer Shopping Trends¶

About Dataset¶

This dataset captures detailed information about customer transactions, providing a comprehensive view of purchasing behavior across various demographics, regions, and product categories. It includes features such as customer demographics (age, gender, location), product details (item purchased, category, size, color, and season), and transaction specifics (purchase amount, payment method, shipping type, and discounts applied).

Data Dictionary¶

Customer ID: Unique identifier assigned to each customer.
Age: The age of the customer (in years).
Gender: The gender of the customer (e.g., Male, Female).
Item Purchased: The specific product bought by the customer (e.g., Blouse, Sneakers, Coat).
Category: The category the purchased item belongs to (e.g., Clothing, Footwear, Accessories, Outerwear).
Purchase Amount (USD): The amount spent by the customer on the purchase (in USD).
Location: The customer's geographical location (e.g., State or region).
Size: The size of the purchased item (e.g., S, M, L, XL).
Color: The color of the purchased item (e.g., Gray, Maroon, White).
Season The season associated with the purchase (e.g., Winter, Spring, Summer, Fall).
Review Rating: The rating provided for the purchased item, typically on a scale (e.g., 1-5).
Subscription Status: Indicates if the customer has an active subscription (e.g., Yes, No).
Payment Method: The mode of payment used for the transaction (e.g., Credit Card, PayPal, Cash, Venmo).
Shipping Type: The shipping option selected by the customer (e.g., Free Shipping, Express, Next Day Air).
Discount Applied: Indicates whether a discount was applied to the purchase (e.g., Yes, No).
Promo Code Used: Indicates whether a promotional code was used during the transaction (e.g., Yes, No).
Previous Purchases: The number of prior purchases made by the customer.
Preferred Payment Method: The payment method most frequently used by the customer (e.g., Credit Card, Venmo, Cash).
Frequency of Purchases: The customer's purchase frequency (e.g., Weekly, Fortnightly, Monthly, Quarterly, Annually).

Installing Dependencies¶

👉Ignore it if already installed

!pip install numpy
!pip install pandas
!pip install matplotlib
!pip install seaborn

In [1]:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

1. Data Understanding¶

Display the dataset's summary statistics.
Check for missing values and handle them if present.
Determine the data types of each column.

In [2]:

df = pd.read_csv(r"C:\Users\harsh\Downloads\shopping_trends.csv")
df.head()

Out[2]:

	Customer ID	Age	Gender	Item Purchased	Category	Purchase Amount (USD)	Location	Size	Color	Season	Review Rating	Subscription Status	Payment Method	Shipping Type	Discount Applied	Promo Code Used	Previous Purchases	Preferred Payment Method	Frequency of Purchases
0	1	55	Male	Blouse	Clothing	53	Kentucky	L	Gray	Winter	3.1	Yes	Credit Card	Express	Yes	Yes	14	Venmo	Fortnightly
1	2	19	Male	Sweater	Clothing	64	Maine	L	Maroon	Winter	3.1	Yes	Bank Transfer	Express	Yes	Yes	2	Cash	Fortnightly
2	3	50	Male	Jeans	Clothing	73	Massachusetts	S	Maroon	Spring	3.1	Yes	Cash	Free Shipping	Yes	Yes	23	Credit Card	Weekly
3	4	21	Male	Sandals	Footwear	90	Rhode Island	M	Maroon	Spring	3.5	Yes	PayPal	Next Day Air	Yes	Yes	49	PayPal	Weekly
4	5	45	Male	Blouse	Clothing	49	Oregon	M	Turquoise	Spring	2.7	Yes	Cash	Free Shipping	Yes	Yes	31	PayPal	Annually

In [3]:

df.describe()

Out[3]:

	Customer ID	Age	Purchase Amount (USD)	Review Rating	Previous Purchases
count	3900.000000	3900.000000	3900.000000	3900.000000	3900.000000
mean	1950.500000	44.068462	59.764359	3.749949	25.351538
std	1125.977353	15.207589	23.685392	0.716223	14.447125
min	1.000000	18.000000	20.000000	2.500000	1.000000
25%	975.750000	31.000000	39.000000	3.100000	13.000000
50%	1950.500000	44.000000	60.000000	3.700000	25.000000
75%	2925.250000	57.000000	81.000000	4.400000	38.000000
max	3900.000000	70.000000	100.000000	5.000000	50.000000

In [4]:

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3900 entries, 0 to 3899
Data columns (total 19 columns):
 #   Column                    Non-Null Count  Dtype  
---  ------                    --------------  -----  
 0   Customer ID               3900 non-null   int64  
 1   Age                       3900 non-null   int64  
 2   Gender                    3900 non-null   object 
 3   Item Purchased            3900 non-null   object 
 4   Category                  3900 non-null   object 
 5   Purchase Amount (USD)     3900 non-null   int64  
 6   Location                  3900 non-null   object 
 7   Size                      3900 non-null   object 
 8   Color                     3900 non-null   object 
 9   Season                    3900 non-null   object 
 10  Review Rating             3900 non-null   float64
 11  Subscription Status       3900 non-null   object 
 12  Payment Method            3900 non-null   object 
 13  Shipping Type             3900 non-null   object 
 14  Discount Applied          3900 non-null   object 
 15  Promo Code Used           3900 non-null   object 
 16  Previous Purchases        3900 non-null   int64  
 17  Preferred Payment Method  3900 non-null   object 
 18  Frequency of Purchases    3900 non-null   object 
dtypes: float64(1), int64(4), object(14)
memory usage: 579.0+ KB

we have non null values

In [5]:

df.dtypes

Out[5]:

Customer ID                   int64
Age                           int64
Gender                       object
Item Purchased               object
Category                     object
Purchase Amount (USD)         int64
Location                     object
Size                         object
Color                        object
Season                       object
Review Rating               float64
Subscription Status          object
Payment Method               object
Shipping Type                object
Discount Applied             object
Promo Code Used              object
Previous Purchases            int64
Preferred Payment Method     object
Frequency of Purchases       object
dtype: object

2. Demographic Analysis¶

Age distribution: Visualize the age distribution of customers.
Gender distribution: Analyze the proportion of male vs. female customers.
Location analysis: Identify the locations with the highest and lowest purchase volumes.

In [6]:

sns.displot(x = df.Age, kde=True , bins= 28)

D:\Installed Softwares\Anaconda\Lib\site-packages\seaborn\axisgrid.py:118: UserWarning: The figure layout has changed to tight
  self._figure.tight_layout(*args, **kwargs)

Out[6]:

<seaborn.axisgrid.FacetGrid at 0x16eeeb06b50>

In [7]:

def agegrp(data):
    if data < 20 :
        return "below 20 "
    elif (data >= 20) & (data < 30):
        return "20-30"
    elif (data >= 30) & (data < 40):
        return "30-40"
    elif (data >= 40) & (data < 50):
        return "40-50"
    elif (data >= 50) & (data < 60):
        return "50-60"
    elif (data >= 60) & (data <= 70):
        return "60-70"
    else:
        return np.nan

In [8]:

d1=df.copy()

In [9]:

d1["age group"]= d1.Age.apply(agegrp)
d1

Out[9]:

	Customer ID	Age	Gender	Item Purchased	Category	Purchase Amount (USD)	Location	Size	Color	Season	Review Rating	Subscription Status	Payment Method	Shipping Type	Discount Applied	Promo Code Used	Previous Purchases	Preferred Payment Method	Frequency of Purchases	age group
0	1	55	Male	Blouse	Clothing	53	Kentucky	L	Gray	Winter	3.1	Yes	Credit Card	Express	Yes	Yes	14	Venmo	Fortnightly	50-60
1	2	19	Male	Sweater	Clothing	64	Maine	L	Maroon	Winter	3.1	Yes	Bank Transfer	Express	Yes	Yes	2	Cash	Fortnightly	below 20
2	3	50	Male	Jeans	Clothing	73	Massachusetts	S	Maroon	Spring	3.1	Yes	Cash	Free Shipping	Yes	Yes	23	Credit Card	Weekly	50-60
3	4	21	Male	Sandals	Footwear	90	Rhode Island	M	Maroon	Spring	3.5	Yes	PayPal	Next Day Air	Yes	Yes	49	PayPal	Weekly	20-30
4	5	45	Male	Blouse	Clothing	49	Oregon	M	Turquoise	Spring	2.7	Yes	Cash	Free Shipping	Yes	Yes	31	PayPal	Annually	40-50
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
3895	3896	40	Female	Hoodie	Clothing	28	Virginia	L	Turquoise	Summer	4.2	No	Cash	2-Day Shipping	No	No	32	Venmo	Weekly	40-50
3896	3897	52	Female	Backpack	Accessories	49	Iowa	L	White	Spring	4.5	No	PayPal	Store Pickup	No	No	41	Bank Transfer	Bi-Weekly	50-60
3897	3898	46	Female	Belt	Accessories	33	New Jersey	L	Green	Spring	2.9	No	Credit Card	Standard	No	No	24	Venmo	Quarterly	40-50
3898	3899	44	Female	Shoes	Footwear	77	Minnesota	S	Brown	Summer	3.8	No	PayPal	Express	No	No	24	Venmo	Weekly	40-50
3899	3900	52	Female	Handbag	Accessories	81	California	M	Beige	Spring	3.1	No	Bank Transfer	Store Pickup	No	No	33	Venmo	Quarterly	50-60

3900 rows × 20 columns

In [10]:

sns.barplot(x= d1["age group"], y=d1.index)

Out[10]:

<Axes: xlabel='age group'>

no significant change in shopping pattern of each age group. thus people of all age shopped regularly .

In [11]:

sns.countplot(x=d1.Gender)

Out[11]:

<Axes: xlabel='Gender', ylabel='count'>

we see that males have purchased more than females.

In [12]:

plt.figure(figsize=(10,6))
sns.countplot(x=d1.Location)
plt.xticks(rotation=90)
plt.show()

We can see highest purchasing from "MONTANA" and lowest purchasing from Rhode Island

3. Purchase Behavior¶

Average purchase amount by category: Compare average spend across product categories (e.g., Clothing, Footwear).
Analyze the frequency of purchases and identify customers with the most frequent purchases.
Payment method trends: Identify the most and least preferred payment methods.

In [13]:

sns.countplot(x=d1.Category, hue=d1.Gender)

Out[13]:

<Axes: xlabel='Category', ylabel='count'>

highest shopped category is "Clothing"

In [14]:

meanprice=d1.groupby("Category")["Purchase Amount (USD)"].mean()
meanprice

Out[14]:

Category
Accessories    59.838710
Clothing       60.025331
Footwear       60.255426
Outerwear      57.172840
Name: Purchase Amount (USD), dtype: float64

In [15]:

totalpurchase=d1.groupby("Category")["Purchase Amount (USD)"].sum()
totalpurchase

Out[15]:

Category
Accessories     74200
Clothing       104264
Footwear        36093
Outerwear       18524
Name: Purchase Amount (USD), dtype: int64

from above, we can conclude that highest shopped category is "Clothing" with total sales of 104264 USD. but the mean sales of "footware" category is high which means more expensive items from footware category were sold with mean being 60 USD.

In [16]:

sns.countplot(x=d1["Frequency of Purchases"])
plt.xticks(rotation=35)
plt.show()