Business Report

on
Austo Motor Company
Sales and Buyer details

PDS Coded Project

Submitted by – Geetakshi Kandpal
Batch - PGPDSBA.O.JULY24.A

1
 List of contents
1. Objective of this Business
problem……………………………………………………………..…………...5
2. Data and its Dictionary
……………………………………………………………………………………………6
3. Data Overview ………………………………………………………………
………………………………………7
4. Univariate Analysis summary.……………………………………………………………….
……………….12
5. Bivariate analysis
summary………………………………………………………………………………......24
6. Multivariate Analysis
summary……………………………………………………………………………….29
7. Key Questions……………………..
……………………………………………………………………………….32
8. Outlier Detection and
Treatment…………………………………………………………………..…………37
9. Actionable Insights and
Recommendations………………………………………………………………38
10.Further Analysis that can be done
…………………………………………………………………………..40
11.References…………………………………….
……………………………………………………………..……..41

2
 List of figures
Basic information about the
dataset…………………………………………………………………………………...7
Statistical summary of the
dataset……………………………………………………………………………………...8
Missing values in
dataset…………………………………………………………………………………………………..9
Values in the dataset after
treatment………………………………………………………………………………….10
Dataset list after
revision………………………………………………………………………………………………….11
Observation on
age………………………………………………………………………………………………………….
12
Observation on
gender……………………………………………………………………………………………………..
13
Observation on
profession………………………………………………………………………………………………..1
4
Observation on marital
status……………………………………………………………………………………………14
Observation on
education…………………………………………………………………………………………………
15
Observation on no. of
dependents……………………………………………………………………………………..16
Observation on personal
loan……………………………………………………………………………………………17

3
Observation on house
loan……………………………………………………………………………………………….17
Observation on Partner
working…………………………………………………………………………………………18
Observation on
salary………………………………………………………………………………………………………
19
Observation on partner
salary……………………………………………………………………………………………20
Observation on total
salary……………………………………………………………………………………………….21
Observation on
price……………………………………………………………………………………………………….2
2
Observation on
make………………………………………………………………………………………………………
23
Correlation by
heatmap……………………………………………………………………………………………………
24
Bivariate analysis of Make with respect to Qualitative
Data……………………………………………………25
Bivariate analysis of Make with respect to Qualitative
Data……………………………………………………26
Bivariate analysis with Pair plots (part-1)
…………………………………………………………………………….27
Bivariate analysis with Pair plots (part-2)
…………………………………………………………………………….27
Bivariate analysis of gender with other
categories………………………………………………………………..28
Price vs Total Salary with respect to the Make of the
car………………………………………………………..29
Price vs Total Salary with respect to
Gender………………………………………………………………………..30
Price vs Total Salary with respect to the no. of
dependents……………………………………………………31
Answer 1: Do men tend to prefer SUVs more compared to
women?.............................................32

4
Answer 2: Likelihood of a salaried person buying a
Sedan………………………………………………………33
Answer 3: What evidence or data supports Sheldon Cooper's claim that a salaried
male is an easier target for a SUV sale over a Sedan
sale?.....................................................................................34
Answer 4 (part-1): Spending with respect to Gender- Total and
Mean…………………………………….34
Answer 4 (part-2): How does the amount spent on purchasing automobiles vary by
gender?.........35
Answer 5: How much money was spent on purchasing automobiles by individuals
who took a personal
loan?..........................................................................................................................3
5
Answer 6: How does having a working partner influence the purchase of higher-
priced cars?........36
Outlier detection and
treatment…………………………………………………………………………………………37

5
 1.Objective of the Business report

1.1 Context:
Analysts are required to explore data and reflect on the insights. Clear writing skills
are an integral part of a good report. Note that the explanations must be such that
readers with minimum knowledge of analytics are able to grasp the insight.
Austo Motor Company is a leading car manufacturer specializing in SUV, Sedan, and
Hatchback models. In its recent board meeting, concerns were raised by the
members about the efficiency of the marketing campaign currently being used. The
board has decided to rope in analytics professional to improve the existing
campaign.

1.2 Objective:
They want to analyze the data to get a fair idea about the demand of customers
which will help them in enhancing their customer experience. Suppose you are a
Data Scientist at the company and the Data Science team has shared some of the
key questions that need to be answered. Perform the data analysis to find answers
to these questions that will help the company to improve the business.

1.3 Plan of Action:

We have conducted a comprehensive analysis on the dataset with the sole aim of
extracting valuable insights for a marketing campaign.
Our approach involves performing extensive data analysis, incorporating
exploratory data analysis (EDA) with univariate, bivariate, and multivariate
techniques, along with utilizing various plots to derive meaningful observations from
the data.

The Analysis will be done as follows-

1) Technical information about the dataset and discrepancies present in data.
2) Univariate analysis for unique attributes exploration drawing insights for
business.
3) Multivariate analysis to understand the relationships among the variables to gain
deeper insights.

6
4) Identify patterns in the buyers groups and purchase history which can help to
develop an improved Marketing strategy.

2.Data and its Dictionary

2.1 Data Description

 Age: The age of the individual in years.
 Gender: The gender of the individual categorized as male or female.
 Profession: The occupation or profession of the individual.
 Marital_status: The marital status of the individual, such as married &,
single
 Education: The educational qualification of the individual Graduate and
Postgraduate
 No_of_Dependents: The number of dependents (e.g., children, elderly
parents) that the individual supports financially.
 Personal_loan: A binary variable indicating whether the individual has taken
a personal loan "Yes" or "No"
 House_loan: A binary variable indicating whether the individual has taken a
housing loan "Yes" or "No"
 Partner_working: A binary variable indicating whether the individual's
partner is employed "Yes" or "No"
 Salary: The individual's salary or income.
 Partner_salary: The salary or income of the individual's partner, if
applicable.
 Total_salary: The total combined salary of the individual and their partner (if
applicable).
 Price: The price of a product or service.
 Make: The type of automobile

7
 3.Data Overview

3.1 Basic Information of the data

The following image shows the basic information about the dataset-

Image 1.1: basic information about the dataset

 All the columns have 1581 observations except Gender and Partner_salary
which has 1528 and 1475 observations respectively indicating that there are
some missing values in them.
 The partner_salary column is being read as a 'float64' data type however it
should be in int (interger) data type. The Age, No_of_Dependents,

8
 Salary,Total_Salary and Price columns are of integer type while the rest of the
columns are object in nature.
 The object type columns contain categories in them

The following image is the statistical summary for the dataset-

Image 1.2: statistical summary for the dataset

 There is a huge difference between indicate quartile and the maximum value
for the Partner_salary and Total_salary indicating that there might be outliers
to the right in these variables.
 The Partner_salary has a wide range indicating that data consists of entries
for different reasons.

3.2 Missing value treatment

The following image shows the missing values in the dataset-

9
 Image 1.3: missing values in the dataset

 The columns Gender and Partner_salary have 53 and 106 missing values in
them.
 There are no missing values in other columns.
After treating the missing values, we observe the following-

10
 Image 1.4: values in the dataset after treatment

We observe from the following image that the data type of

the Partner_salary column has been successfully changed to integer data type.

11
Image 1.5: Datatype list after revision

12
 4.Univariate Analysis

4.1 Observation on Age

Image 2.1: Observation on Age

 The distribution of Age is right skewed.
 There are no outliers in this variable.

13
 We can infer from the plot that the people between the age of 25-30 years
have purchased the car.

4.2 Observation on Gender

Image 2.2: Observation on Gender

 We can infer from the above plot that the majority of the buyers are males.

4.3 Observation on Profession

14
 Image 2.3: Observation on Profession

 There are more buyers who are salaried as compared to being businessmen.

4.4 Observation on Marital Status

Image 2.4: Observation on Marital Status

 Mostly the people buying the car are married.

15
4.5 Observation on Education

Image 2.5: Observation on Education

 People who are post graduate are buying more cars in comparison to the
graduates.

4.6 Observation on Number of Dependents

16
 Image 2.6: Observation on Number of dependents

 We observe that the No_of_Dependents has a normal distribution.

 There are outliers present in this data.
 Mostly there are 2 or 3 number of dependents.

4.7 Observation on Personal loan

17
 Image 2.7: Observation on personal loan

 Half the people buying the car have taken a personal loan.

4.8 Observation on Home loan

Image 2.8: Observation on Home loan

18
  Lesser buyers have taken a home loan, indicating a financially stable group.

4.9 Observation on Partner working

Image 2.9: Observation on Partner working

 There are a greater number of the buyers who have working partners.

4.10Observation on Salary

19
 Image 2.10: Observation on Salary

 There is an extensive range of salaries, with a few buyers who are higher
earners.

4.11Observation on Partner Salary

20
 Image 2.11: Observation on Partner Salary

 We observe that some partners are earning a significant income whereas

many do not contribute financially. The zero values could represent home-
maker partners or partners not currently working.

4.12Observation on Total Salary

21
 Image 2.12: Observation on Total Salary

 The total salary is the sum of the buyer's as well as their partner's earnings,
hence it shows a wide variety, with some families earning a lot more than
others.
 The presence of outliers suggests some of buyers are earning significantly
more than others, which could skew averages.

22
4.13Observation on Price

Image 2.13: Observation on Price

 The above analysis suggests a wide range in the prices of the cars.

4.14Observation on Make

23
 Image 2.14: Observation on Make

 There are three unique models, with "Sedan" being the most frequent.

5. Bivariate Analysis
24
5.1 Correlation by Heatmap

Image 3.1 Correlation by Heatmap

 The positive correlation between Salary and Age suggests that salary tends to
increase with age, which is normal in many professions as the experience
grows.
 A low correlation between Age and Partner salary might indicate that age
isn't a strong determinant of salary in this dataset.
 There is a high correlation between Age and Price which could indicate that
older individuals are associated with higher prices since they have more
purchasing power.
 There is a moderate correlation between Price and Salary indicating there
might be other factors a family takes into consideration before purchasing.
 As expected, Age is negatively correlated with No_of_Dependents. If the age
is less, then no of dependents on the individual will also be less. This is
aligned with our intuitive understanding.
 A low correlation between Age and Salary might suggest that age doesn’t
significantly impact salary in the dataset.
 Total_salary is highly correlated with Salary and Partner salary; it makes
sense because Total_salary is a sum of them both.
 It is important to note that correlation does not imply causation.

25
 There does not seem to be a strong relationship between Price and Partner
salary.

5.2 Bivariate analysis of Make with respect to

Qualitative Data

Image 3.2 Bivariate analysis of Make with respect to Qualitative Data

 Gender: Males usually prefer Sedan and Hatchback; Females prefer SUV and
Sedan more, not prefer Hatchback
 Profession: Similar choice for make of car, for salaries, Sedan> Hatchback>
SUV. Business persons prefer sedan and hatchback equally.
 Marital_status: Married persons prefer Sedan whereas single people prefer
Hatchback, do not prefer SUV.
 Education: Similar choice of Post graduate and Graduates; Sedan is the most
preferred make.
 Personal_loan: SUV is the least preferred however Sedan most preferred.
 House_loan: SUV not preferred by persons with House loan and Sedan most
preferred by both.
 Partner_working: SUV is the least preferred whereas Sedan is the most
preferred model.

5.3 Bivariate analysis of Make with respect to

Quantitative Data

26
 Image 3.3 Bivariate analysis of Make with respect to Quantitative Data

 The make preference changes with Age - Younger age group (18-30) prefer
the Hatchback model whereas the age group of 28-45 prefer Sedans and the
age group of 30-55 prefer SUVs.
 Almost 75% of total sales for each make is made to customers with 1-3
dependents, Sedan being more preferred by customers with 1-2 dependents.
The single people do not prefer sedans, they either go for SUVs or
Hatchbacks.
 Higher salary group of people prefer SUVs whereas lower salary group prefers
Hatchbacks.
 Buyers with low Partner salary prefer to go for Hatchbacks as it is evident
from the low median for hatchbacks.
 SUVs are the most expensive make followed by Sedans and then Hatchbacks.

5.4 Bivariate analysis with Pair plots

27
 Image 3.4: Bivariate Analysis with pair plots (part-1)

Image 3.5: Bivariate Analysis with pair plots (part-2)

 We get the same insights as from the correlation plot

 There does not seem to be a strong relationship between number of
dependents and Total salary.
 Again, we observe that there is no correlation between Price and Total salary.

28
5.5 Bivariate analysis of Gender with other
categories

Image 3.6: Bivariate analysis of gender with other categories

 Most of the car owners belong to the age group of 20-30 and are mostly
males.
 Majority of young males prefer low price cars with the Hatchbacks model.
 Most female car owners are of 30 or more age group.

29
 6. Multivariate Analysis

6.1 Price versus Total Salary based on the Make of the

car

Image 4.1: Price vs Total Salary with respect to the Make of the car

 Higher income groups of buyers generally prefer higher priced SUVs.

 Mid Salaried people prefer Sedan.
 A Hatchback is preferred by people with comparatively low average salary as
it is easier on the pocket.
 Total salary is considered while analyzing this.

6.2 Price versus Total Salary with respect to Gender

30
 Image 4.2: Price versus Total Salary with respect to Gender

 We observe that there are a lot of the male buyers in the lower to middle
income group buying cars that are lower priced.
 There are few female buyers for the lower priced cars.
 Majority of the female buyers prefer moderately to high priced cars.

6.3 Price versus Total Salary based on the Number of

dependents

31
 Image 4.3 Price vs Total salary with respect to the number of dependents

 The buyers with no dependents prefer buying low to mid-range cars.

 The buyers with only single dependent prefer buying mid-range cars.
 There are buyers with 3-4 dependents with total salary significantly above
the general trend might indicate outliers who purchased a very expensive.

7. Key Questions
32
7.1 Question 1: Do men tend to prefer SUVs more
compared to women?

Answer1: The female buyers tend to prefer SUVs more as compared to men. There
are 53 % of female buyers who have purchased SUVs whereas only 10% of male
buyers for SUVs. The below given graph shows the tendency of the purchase of
SUVs amongst the different genders.

Image 5.1: Answer 1: Do men tend to prefer SUVs more compared to women?

33
7.2 Question 2: Likelihood of a salaried person buying a
Sedan

Answer 2: The likelihood of a salaried person purchasing a sedan is approximately

42.8%. The pie chart shown below visualizes this, showing the proportion of salaried
individuals who have purchased a sedan versus those who have not.

Image 5.2: Answer 2: Likelihood of a salaried person buying a Sedan

7.3 Question 3: What evidence or data supports Sheldon

Cooper's claim that a salaried male is an easier target for
a SUV sale over a Sedan sale?

Answer: To evaluate the claim that a salaried male is an easier target for an SUV
sale compared to a sedan sale, we compared the number of SUVs and sedans
purchased by salaried males in the dataset. The number of SUVs purchased by

34
salaried males is 85, whereas number of Sedans purchased by salaried males is
279. We calculated the relevant statistics and then plotted the comparison in the
bar graph given below.

Image 5.3: Answer 3: What evidence or data supports Sheldon Cooper's claim that a
salaried male is an easier target for a SUV sale over a Sedan sale?

7.4 Question 4: How does the amount spent on

purchasing automobiles vary by gender?

Answer 4: We observe that the total Spending on Automobiles by Gender shows a

plot summarizing the spending behavior (in terms of Price) for each gender group in
this dataset. We notice that the mean spending of the females is higher than males.
Even though the female buyers participating in the dataset are less, however, we
see that 50% are spending more as compared to their male counterparts.
We also observe outliers in the data, which might indicate that a lot of male buyers
might have participated in the survey for this dataset.
Hence, the below given plot visualizes these differences, with the Male group having
a significantly higher total spending, while the Female group has a higher average
spending per individual.

35
 Image 5.4: Answer 4 (part-1): Spending with respect to Gender- Total and Mean

Image 5.5: Answer 4 (part-2): How does the amount spent on purchasing
automobiles vary by gender?

7.5 Question 5: How much money was spent on purchasing

automobiles by individuals who took a personal loan?

Answer 5: The total amount spent on purchasing automobiles by individuals who

took a personal loan is $ 26644000

36
Image 5.6: Answer 5: How much money was spent on purchasing automobiles by
individuals who took a personal loan?

7.6 Question 6: How does having a working partner

influence the purchase of higher-priced cars?

Answer 6: Having a working partner does not really influence the purchase of
higher- priced cars.
After determining what qualifies as a "higher-priced car" (i.e., top 25% of prices), we
compared the proportion of higher-priced car purchases between individuals with a
working partner versus those who did not and plotted the results which are given
below.

37
Image 5.7: Answer 6: How does having a working partner influence the purchase of
higher-priced cars?

8. Outlier Detection and Treatment

Let's visualize all the outliers present in data :

38
 Image 6.1: Outlier Detection and Treatment

 The Total salary column has a wide range of values with lots of outliers.
However, we are not going to treat this column since the total salary can
have a varying range and we can miss out on some genuine values if we
treat this column.
 The No of Dependents column have some outliers, however all of them seem
to be genuine values, so we are not going to treat the outliers present in
these columns.

9.Actionable Insights and Recommendations

We analyzed a dataset of more than 1500 automobile buyer information, for

different ages, salaries, no of dependents and makes. The data spanned over

39
different Price ranges of the cars. The main feature of interest here is the pattern
that can be seen in the buyer groups and purchase history which can help us
develop an improved Marketing strategy.

9.1 Insights
We have been able to conclude that -

1) HATCHBACKS are primarily preferred by single males or married couples within

the age group of 20 to 30 years which is also interestingly the largest group based
on data and has a wide Total salary range spanning between 20k to 120k.
This suggests a need of reinforced marketing campaign for Hatchbacks focusing on
the young adults and young couples.

2) SEDAN has the widest Price range between $20k to $60k that coincides with the
Hatchback and SUV segments also. Married couples with low total Salary range
around $30K prefer only low-priced Sedans or Hatchbacks. Singles in the age group
between 30 to 45 years prefer costlier variant greater than $30k and not the basic
Sedan variant. Hence, for the youth, the Hatchback model and lower price tag
Sedans can be marketed.
Therefore, for the youth, Hatchbacks and entry level Sedans can be marketed.

3) SUVs are the preferred choice of all age groups between 28 to 55 years. Amongst
the SUV owners, the married males go for higher end, that is greater than $50000
SUV variant whereas female SUV owners prefer lower end SUV variants, that is
lesser than $50000. SUV owners either married or single greater than 45years of
age prefer only higher end SUV variants.
Hence for SUVs, the marketing campaigns can be focused more on male car owners
for higher variant SUVs.

9.2 Recommendation to Business

1. Develop targeted marketing strategies for customers with personal loans.
2. Design products or services tailored made to the high spending premium
segments.

40
3. Implement risk assessment models that consider loan status and spending
behavior.
4. Gather additional data on customer preferences, feedback, and behaviors to
enrich the dataset. This can provide deeper insights on the marketing
campaigns and improve decision making.
5. Use insights from their spending patterns to create personalized offers and
promotions.

10. Further Analysis that can be done

41
The following are some of the suggestions that come to our mind regarding the
further analysis that can be done to strengthen the marketing campaign:

1. Investigate how the spending behavior of the buyer or other variables change
over time. This can help in understanding trends and making forecasts.
2. Construct predictive models to forecast future spending.
3. Create predictive models to forecast various loans and how it affects the
sales.
4. Explore customer behavior in detail, including frequency of purchases of
automobiles and loyalty towards the model of the car.
5. Scrutinize customer feedback and reviews to measure the sentiment and
identify areas for improvement.

11. References

42
 Great Learning video lectures by Daniel A Mitchell, Clinical Assistant
Professor, McCombs School of Business, UT Austin
 Great Learning Mentored Learning by Netali Agarwal

43