Scribd
Upload
17 views4 pages

Customer Behavior Analysis in Python

Uploaded by

Abdulaziz B Umar
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
17 views4 pages

Customer Behavior Analysis in Python

Uploaded by

Abdulaziz B Umar
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Assignment-3

Business Analytics

Submission:
You need to submit the python code and the screenshots of the execution results
for every question. Please submit a pdf file and a python file.
The use any python IDE .
Problem Context:
Customer behavior refers to an individual's buying habits, including social trends,
frequency patterns, and background factors influencing their decision to buy
something.
Businesses study customer behavior to understand their target audience and create
more-enticing products and service offers. It helps a business to better understand
its customers and makes it easier for them to modify products according to the
specific needs, behaviors and concerns of different types of customers.
Customers are first segmented into buyer personas based on their common
characteristics. Then, each group is observed at the stages on your customer journey
map to analyze how the personas interact with your company. For example, instead
of spending money to market a new product to every customer in the company’s
database, a company can analyze which customer segment is most likely to buy the
product and then market the product only on that particular segment.
Goal: Shape the data to create different customer segments based on their
respective persona.
Data : You are provided [Link]
The data has many columns organized based on their scope as follows:
People
• ID: Customer's unique identifier
• Year_Birth: Customer's birth year
• Education: Customer's education level
• Marital_Status: Customer's marital status
• Income: Customer's yearly household income
• Kidhome: Number of children in customer's household
• Teenhome: Number of teenagers in customer's household
• Dt_Customer: Date of customer's enrollment with the company
• Recency: Number of days since customer's last purchase
• Complain: 1 if the customer complained in the last 2 years, 0 otherwise
Products
• MntWines: Amount spent on wine in last 2 years
• MntFruits: Amount spent on fruits in last 2 years
• MntMeatProducts: Amount spent on meat in last 2 years
• MntFishProducts: Amount spent on fish in last 2 years
• MntSweetProducts: Amount spent on sweets in last 2 years
• MntGoldProds: Amount spent on gold in last 2 years

Promotion
• NumDealsPurchases: Number of purchases made with a discount
• AcceptedCmp1: 1 if customer accepted the offer in the 1st campaign, 0 otherwise
• AcceptedCmp2: 1 if customer accepted the offer in the 2nd campaign, 0
otherwise
• AcceptedCmp3: 1 if customer accepted the offer in the 3rd campaign, 0 otherwise
• AcceptedCmp4: 1 if customer accepted the offer in the 4th campaign, 0 otherwise
• AcceptedCmp5: 1 if customer accepted the offer in the 5th campaign, 0 otherwise
• Response: 1 if customer accepted the offer in the last campaign, 0 otherwise

Place
• NumWebPurchases: Number of purchases made through the company’s website
• NumCatalogPurchases: Number of purchases made using a catalogue
• NumStorePurchases: Number of purchases made directly in stores
• NumWebVisitsMonth: Number of visits to company’s website in the last month

Part I: Data Overview


1. Use the pandas library to read the data file and to create the data frame.
2. Display the first 5 rows and the last 3 rows of your data.
3. Show quick statistics of your data
4. Show the data type of each column
5. Show how many columns and rows in your data.
6. Show the list of columns in your data frame.
7. Show the number of duplicated rows in your data.

Part II: Data Preparation and Cleaning


8. Rename the columns belonging to Products by removing Mnt from them.
a. You should have: 'Wines','Fruits','Meat','Fish','Sweet','Gold'
9. Rename the columnsbelonging to Place by removing Num and Purchases from
them.
a. You should have: 'Web ','Catalog', 'Store'
10. Keep only NumDealsPurchases column in the Promotion section
11. Delete the column NumWebVisitsMonth
12. The attribute Year_Birth is not helpful to segment your data. Instead create a
new derived column named Age.
13. Create a new column named Spending that holds the total amount spent on all
products categories.
14. Change the Dt_Customer type from object to datetime. Use the format YYYY-
MM-DD.
15. Show the unique values of the column data Marital_Status.
16. It is better to work with lower categorical values. Hence, we’ll classify customers
in two segments for their martial status: change the column Marital status as
follows: a. Change the values: Divorced, Single, Absurd and Widow to 'Alone'
b. Change Married and Together to 'Not Alone
17. Show the unique values of the column Education.
18. We want to segment the customers in 2 groups based on their education level:
change the column Education as follows: a. Change the values: 'Basic' and '2n
Cycle' to 'Undergraduate',
b. All other values to 'Postgraduate'
19. Have a look at column Income. Investigate the presence of outliers? Delete them
accordingly.
20. Show quick statistics for the column Income.
21. Show the number of missing values for each column.
22. Fill the missing values with the average.
23. Create a new column named Children to hold the total number of children for
every customer.
24. Create a new column Has_Complaint as follows: a. 'Has complaint' if the
customer complained in the last 2 years,
b. Otherwise 'No complaint'
25. Show the first 5 rows of customers with ‘No complaint’

Part III: Data Visualization & Analytics


26. Let’s analyse the profile of customers based on their background factors. To do
so, we will use our cleaned data frame (in part II) to create a new one with the
columns: Age, Spending, Marital_Status, Education, Income, Children,
Dt_Customer and Has_Complaint.
27. Create histogram to show the number of children per customers in your data.
What do you see?
28. Create a histogram and a Boxplot to show the Income of customers. What do
you find?
29. Create a histogram and a Boxplot to show the Spending of customers. What do
you find?
30. Create a histogram and a boxplot to analyse the customers based on their Age.
31. Create a histogram to analyse the Education level of the company’s customers.
What do you find?

**The column Education is defined as Object and has 2 values (Undergraduate,


Postgraduate). You need to shape the values as numeric to be able to plot the
histogram.

Common questions

Powered by AI

Data preparation and cleaning are crucial steps in customer behavior analysis as they ensure data accuracy and reliability. By renaming columns for clarity, handling missing values, converting data types appropriately, and removing outliers, the dataset becomes more reflective of true consumer behavior patterns, leading to more valid insights and conclusions .

Histograms and boxplots provide a visual summary of customer data distributions, revealing patterns such as central tendency, variability, and outliers. For income and spending data, these visualizations help in understanding the distribution, highlighting trends, median values, and potential outliers, which are critical for formulating data-driven business decisions and refining customer segmentation models .

Simplifying marital status and education levels into broader categories reduces complexity and enhances analytical clarity. By grouping similar values, analysts can identify larger patterns and trends across the dataset, making it easier to derive insights that are impactful for strategic decision-making and marketing .

'Spending' is a key indicator of customer purchasing behavior, reflecting how much customers invest in the products offered. It should be calculated by summing the amounts spent across all product categories, providing a comprehensive view of each customer's spending habits and allowing for further segmentation and targeting based on financial engagement with the company .

Analyzing customer complaints helps identify areas of dissatisfaction, guiding improvements in products or services. By segmenting customers based on 'Has_Complaint', businesses can tailor their service and resolve specific issues, ultimately improving customer satisfaction and loyalty, as these insights drive more effective customer service strategies .

Segmenting customer personas allows a business to tailor marketing strategies specifically to the characteristics and behaviors of each segment, optimizing resource allocation and improving targeting efficiency. By analyzing which customer segment is most likely to purchase a product, businesses can focus their marketing efforts on those groups, thereby increasing the likelihood of successful conversions and improving overall marketing efficiency .

Adjusting 'Dt_Customer' from object to datetime facilitates temporal analyses, such as customer tenure with the company. This conversion allows for the usage of date-specific analytical functions within data science tools, enhancing the ability to conduct time series analysis, segment customers by enrolment age, and study the correlation between engagement time and purchasing behavior .

Detecting and removing income outliers is crucial as they can skew analysis results and lead to inaccurate insights into customer's financial behaviors. By removing these anomalies, the analysis reflects a more accurate representation of the typical customer, enhancing predictive models and improving marketing strategy formulation targeting different income segments .

Transforming 'Year_Birth' into 'Age' makes data analysis more intuitive and directly applicable to understanding customer demographics. It simplifies age-related queries and visualizations, allowing for straightforward segmentation and analysis of age-dependent buying behaviors and trends .

Visualizing customer education levels through histograms allows businesses to assess the distribution of educational backgrounds among their customers. This insight aids in identifying target segments that align with specific educational profiles, which can be targeted with customized marketing strategies that resonate with their level of education and likely preferences .

You might also like