Instructions
Requirements and Specifications
| Column | Description |
| product_account_id | user identifier specific and unique to each user Primary and unique key in user table |
| registration_date | date that the user signed up for FanDuel |
| first_deposit_date | date the user made their first deposit on the FanDuel site/app |
| first_paid_entry_date | date the user made their first paid entry on the FanDuel site/app |
| first_deposit_amount | dollar amount of the users first deposit Relates to first_deposit_date |
| registration_paid_media_partner_name | media partner that is associated with driving the user's registration. All partners are paid media partners aside from "ORGANIC" |
| first_deposit_paid_media_partner_name | media partner that is associated with driving the user's first deposit. All partners are paid media partners aside from "ORGANIC" |
| first_paid_entry_paid_media_partner_name | media partner that is associated with driving the user's first paid entry. All partners are paid media partners aside from "ORGANIC" |
| registration_state | the state the users was in when they signed up for FanDuel. Relates to registration_date |
| Column | Description |
| product_account_id | user identifier specific and unique to each user. Is not unique on the entry table |
| entry_week | the week a user entered a contest. Week is noted as the first day of the week |
| entry_sport | the sport associated with the user's entry |
| contest_type | the type of contest associated with the user's entry |
| entry_source | the platform associated with the user's entry |
| weekly_entry_fee | the summed total of an users weekly entries |
| weely_total_entry_winnings | the summed total of an users weekly winnings represented as a gross total. To find net winnings you would do "weekly_entry_fee" - "weekly_total_entry_winnings" |
| total_entries | the total count of entries a user made in a week |
| avg_contest_size | the average size of the contest the users entered into based on the combination of "entry_sport", "contest_type", and "entry_source" in a week |
- Given the above data sets show what month and year (Eg. March 2019, April 2018, etc.) has the largest number of registrations? Please perform this using SQL and provide both the output and your query to produce your results.
- select count(registration_date) as countofregistrationdate ,registration_date
- from user_data
- group by registration_date
- order by countofregistrationdate DESC
- limit 1
- Given the above data sets show the top 10 ad partners by total users per First Paid Entry Credit (column “first_paid_entry_paid_media_partner_name”). Please perform this using SQL and provide both the output and your query to produce your results. Be sure to not include users who have not made a first paid entry.
- Like the above show the top 10 Ad Partners, grouped by First Paid Entry Credit (column “first_paid_entry_paid_media_partner_name”) and ranked by total entry fees for users who first paid entry data was in September. In addition to total entry fees, include the total count of entries and the average entry fee amount per user per ad partner. Please perform this using SQL and provide both the output and your query to produce your results.
- For users who registered in 2019, which month, based on a user’s first_paid_entry_date, saw the largest number of unique users make an NBA entry at any point in time? Please perform this using SQL and provide both the output and your query to produce your results.
- What were the top 5 sports in both 2018 and 2019 based on total entries for users that made their first paid entry within one week of registering? Include the average first_deposits_amount, and average time between registration and first paid entry per sport and year as well. Please write this as a single query using the “rank()” function. Please perform this using SQL and provide both the output and your query to produce your results. Note, we want to know the top 5 sports for each year not overall in 2018 and 2019 combined.
Source Code
-- question 1
select count(*) as 'Count', DATE_FORMAT(registration_date, '%M %Y') as 'Date'
from user
group by DATE_FORMAT(registration_date, '%M %Y')
order by count(*) DESC
LIMIT 1
-- question 2
select count(*) 'Count', first_paid_entry_paid_media_partner_name
FROM user GROUP BY first_paid_entry_paid_media_partner_name
ORDER BY count(*) DESC
LIMIT 10;
-- question 3
select B.total_entries 'Total Entries', B.weekly_entry_fee 'Total Entry Fees',
(B.weekly_entry_fee/B.total_entries) 'Average Entry Fee', MONTH(B.entry_week) 'Month',
A.first_paid_entry_paid_media_partner_name 'Partner Name', A.product_account_id 'User ID'
FROM user A
INNER JOIN entry B
ON A.product_account_id = B.product_account_id AND MONTH(B.entry_week) = 9
LIMIT 10;
-- question 4
SELECT YEAR(A.registration_date) 'Registered at Year', MONTH(A.first_paid_entry_date) 'Paid at Month',
B.entry_sport 'Entry Sport', COUNT(DISTINCT A.product_account_id) 'Number of users'
FROM user A
INNER JOIN entry B
ON B.entry_sport = 'NBA'
WHERE YEAR(A.registration_date) = 2019
GROUP BY MONTH(A.first_paid_entry_date)
ORDER BY COUNT(A.product_account_id) DESC
LIMIT 1
-- question 5
SELECT YEAR(A.entry_week) 'Year', A.total_entries 'Total Entries', A.entry_sport 'Sport',
AVG(DATEDIFF(B.first_paid_entry_date, B.registration_date)) 'Average time between registration and first paid entry',
AVG(B.first_deposit_amount) 'Average first deposit amount',
RANK() OVER (PARTITION BY A.entry_sport ORDER BY A.total_entries) 'Rank'
FROM entry A
INNER JOIN user B
ON A.product_account_id = B.product_account_id
WHERE DATEDIFF(B.first_paid_entry_date, B.registration_date) <= 7
AND A.entry_sport != ''
AND (YEAR(A.entry_week) = 2018 OR YEAR(A.entry_week) = 2019)
GROUP BY YEAR(A.entry_week), A.entry_sport
ORDER BY YEAR(A.entry_week), 'Rank' ASC