Exercise 1

There’s no particular right/wrong answer in this exercise. Our aim is simply to simulate the type
of work that you might do in your role if you were to join the team, and to see how you approach
it.

Scenario
● You’re a Data Analyst, and one of your duties is to surface useful insights about JET’s
delivery business to help senior leadership make better informed business decisions.
● You’ve been given the enclosed data set containing a variety of data points pertaining to
individual deliveries completed on the JET platform over a specific period of time.
● Your audience (senior leadership) are not experts in the logistical intricacies of our
business. Assume that whatever you tell them will be new information.

Objective
● Extract whatever insights from the data set that you think could be useful to your
audience and summarize them in a brief report using whatever data visualisation tool
you are most comfortable with.
● Where possible, make recommendations based on your findings.

Parameters
● Make any assumptions that you want; just state them somewhere in your finished
product. (And don’t worry about your assumptions being way off -- that’s to be expected,
given that the subject matter is new to you.)
● Graphic design isn’t the focus here, so don’t worry too much about making a work of art.
The emphasis here is on substance over style.
● Please don’t spend too much time on this. You could probably sink many hours into
mining the data set. The goal is to see how you think/work, not to test whether you can
find every meaningful pattern in the data. (Think 2-4 hours’ effort max.)

How to read data sample

● Order_number - unique identifier for order
● Courier_id - unique identifier for courier who delivered order
● Restaurant_id - unique identifier for restaurant who prepared food
● Courier_shift_id - unique identifier for the shift the courier was on when delivering the
order
● Delivery_date - delivery date of the order
● Delivery_placed_time - the time that the customer placed the order on the platform
● Order_assigned_to_courier_time - the time that the courier was assigned to the order
● Order_accepted_by_courier_time - the time that the courier swiped to accept the order
● Courier_in_transit_to_restaurant_time - the time that the courier swiped they were in
transit to the restaurant
● Courier_arrived_at_restuarant_time - the time that the courier swiped that they arrived
at the restaurant
 ● Courier_expected_at_restaurant_time - the time that the courier was expected to
arrive at the restaurant
● Courier_collected_from_restaurant_time - the time that the courier swiped that they
collected the food from the restaurant
● Courier_in_transit_to_customer_time - the time that the courier swiped that they were
in transit to the customer
● Courier_arrived_at_customer_time - the time that the courier swiped that they were
parked at the customer
● Delivery_expected_completed_time - the time that the courier was expected to deliver
the food to the customer
● Delivery_completed_time - the time that the courier swiped that they delivered the food
to the customer
● Courier_held_by_customer_sec - time in seconds that the courier was delayed by the
customer in delivering the food
● Courier_held_by_restaurant_sec - time in seconds that the courier was delayed by the
restaurant in collecting the food
● Distance_courier_assigned_to_restaurant_meters - the distance the courier was
from the restaurant when they were assigned the order
● Distance_restaurant_to_customer_meters - the distance from the restaurant to the
customer
● Courier_assigned_to_restaurant_driving_time_min - estimated time for courier to
travel from where they were assigned order to the restaurant
● Restaurant_to_customer_driving_time_min - estimated time for courier to travel from
the restaurant to the customer

Exercise 2
Please refer to table structures on the following page and complete as many of the questions
below as you can. Keep in mind:

● You can use any variant of SQL

● Don’t worry about getting the syntax exactly correct - focus on the query structure!
● No calculation is required; we’re just looking for the SQL code that would return the
specified value / result set

Questions:
1. Write a query to identify the total number of restaurants in the Restaurant table
2. Write a query to find the earliest transaction date in the Transaction table
3. Write a query that calculates the total revenue (i.e. transaction_value) of all historic
transactions to date
4. Write a query that calculates the average transaction value for each restaurant
brand in the last 7 days
5. Write a query that returns a list of all restaurant names in the database, along with
the number of transactions they processed yesterday. The result set should include
all restaurants regardless of whether or not they had a transaction
6. Write a query that returns a daily count of the number of "Indian" cuisine type
transactions and the number of "Pizza" cuisine type transactions. Your result set
should have one row per transaction date, and the cuisine transaction counts should
be presented in separate columns
7. Write a query that returns the total number of customers whose first ever transaction
was in January 2017
8. Write a query that returns the total number of transactions that a customer makes in their
first 90 days. Your result set should have one row per Customer ID

Transaction table
 ● Contains all historic customer transactions, each of which has a unique Transaction ID.

Table name = “transaction”

Example:

Restaurant table
● Lookup table of all restaurants on the platform, each of which has a unique Restaurant
ID.

Table name = “restaurant”

Example: