SENIOR RISK DATA ANALYST

1/ [SQL]
Assuming that your company is running a loan marketplace where people who want to borrow
money are matched with appropriate loan products provided by different banks, the data
schemas are shown below:

Banks
Column name Data type Notes

bank_id int Primary key

bank_name string Examples: “HSBC”, “Ocean Bank”,...

Products
Column name Data type Notes

product_id int Primary key

loan_amount int The currency unit is USD.

interest_rate float

accepted_risk_level string “low” / “medium” / “high”

bank_id int The bank that provides this product

Customers

Column name Data type Notes

customer_id int Primary key

customer_name string Example value: “Morgan Freeman”

customer_age int Example value: 65

estimated_risk_level string “low” / “medium” / “high”

source string The source that brings this customer to

marketplace

created_date DateTime

Leads
Column name Data type
Notes

customer_id int

product_id int
 apply_date DateTime

​
A customer with an estimated risk level X will only be matched with a product that accepts
risk level X.

Based on the above data tables, please write SQL queries to:

a)​ Show the number of products available for each accepted risk level.

b)​ Show the average interest rates of products provided by HSBC and Techcombank banks.

c)​ Show 2 banks that have the most high-risk products.

d)​ Show which source brings to the marketplace more low-risk customers.

e)​ Show all months of the year 2018 that the number of customers applying for loans is 20%
higher than the monthly average number of customers of the year.

f)​ Show the names of all leads who applied in 2019 and are older than 95% of all leads
who applied in 2017
2/ [BUSINESS]
You are an Analytic Engineer for a company which produces a new generation of electric men
razor. Your company registered an e-commerce site at www.Coolmen-Coolrazors.com 1 month
ago to sell its product online instead of the traditional supermarket channel. During the last
month, it piloted advertising on 2 channels:

● Email Channel

● SMS Channel

Data are extracted from a centralized database and stored in the attached file called
“mkt_data.csv”. Dataset

The schema for this dataset is as follow:

send_date Format: data, date when SMS/Email was sent

estimated_age Format: Integer, ranging from 0 to 100

age_range Format: string. The audience is divided into 4 age ranges

channel Format: string, either SMS or Email

coupon Format: float, the value of coupon expressed in each message, valid for
up to 3 units for each order

clicked Format: binary, either 0 (customer doesn’t click on the link in SMS/Email)
or 1 (they clicked)

last_step Format: string. It can have one of the following values: “received”,
“bounced”, “saw review”, “added to cart”, “payment page”,
“purchased”

nb_units
Format: integer, representing the number of units of customers’ order.

order_value
Format: float, representing the value of the order the customer made.
Already minus the coupon applied.

The column “last_step” is the final point of contact with customers before they leave our
website. Its values are explained below:

● Received: sms/email sent successfully, but not clicked.

● Saw review: scroll down and read the review and information of the product

● Added to cart: customers added the product to cart to check out

● Payment page: they stopped at payment without finishing it

● Purchased: they made an order

Financial Information

Together with the data above, you have additional information about the production cost and
the marketing campaigns.
● The production cost for each razor is 18$.

● Cost per one SMS is $0.050, cost per one email sent is $0.075.

● Each email or SMS will be supplied a coupon that can have a value of 2$, 4$ or 6$.
The coupon is valid for up to 3 razors in each order. They have the option to wrap the
items as a gift. Ignore wrapping and shipping costs.

● The price without coupon is 40$ / razor.

● From experience (and some models), potential customers are divided into 4 age
groups:

○ 18 - 30

○ 31 - 45

○ 46 - 60

○ 60 +

Question

2.a.

For the next quarter, your marketing department has a budget of $60,000 to spend on online
campaigns. How would you allocate it between SMS and Email? Assume that we have a
potential customer pool for each age group as below:
Age Group Pool size

18 - 30
300,000

31 - 45
350,000
 46 - 60 500,000

60+ 200,000

2.b.
Now assume that you are also responsible for the operation of the company’s website. Do you
have any comments or suggestions so that we can improve the website’s performance in
order to maximize net profit?
3/ [Python]
Please find attached the file “messy.xlsx”, use Python or your programming language of choice
to do the following:

● Clean the names of columns to lowercase separated by “_”, remove any empty
column if necessary.

● Change the date column to the same format ‘YYYY-MM-DD’.

● Change the name column to the title case (e.g: Jason Mraz).

● Make a new “email” column with the form:

● Change the phone number column to the format “84……”

● Find any duplicated ID and remove those who join later.

● Filter those who join since 2019 and export to a csv file, delimited by “|”, file name
“emp_{report_date}.csv” with report_date = today.