Challenge requirements

You product manager is reaching out to you and kindly requests some basic report on listings and leads on our platform. Ideally the report is a web-ui so the product manager can look into it on a regular basis.

The report should include:

• Average Listing Selling Price per Seller Type
• Distribution (in percent) of available cars by Make
• Average price of the 30% most contacted listings
• The Top 5 most contacted listings per Month

Milestones

The present state of this repository delivers the following milestones:

1. A web application that displays / outputs the reports which are server genereted html using Express. The style was kept quite simple with the use of Bootstrap.
2. An API endpoint which exposes the data in a structured format. This could be used from other teams within the same company that would like to re-use the aggregations.

Terms & Taxonomy

• Listing: that's the representation of a car being sold.
• Make: "Producer" of cars, for instance BWM, Audi, VW, etc.
• Seller Type: What kind of seller is behind that listing. If it's "private", that means the car will be sold by a private person. If it's dealer, it's means that the car is sold on a dealer shop. "Other" means another types of sellers that ate not relevant in this context.
• Lead: Contact between a car buyer and car seller.

Acceptance Criteria

Average Listing Selling Price per Seller Type

• There are three Seller Types: private, dealer and other.
• For each of these types, it should be provided an average selling price.
• The average price should be formatted as € #,-
• Output format is free for you to choose, but an example could be:

Seller Type	Average in Euro
private	€ 2.500,-
dealer	€ 3.529,-
other	€ 1.200,-

Percentual distribution of available cars by Make

• For each make, it should be reported the percentual amount of listings.
• The report should be sorted by distribution, where makes with biggest numbers stays on top.
• Output format is free for you to choose, but an example could be:

Make	Distribution
Audi	55%
BMW	35%
VW	10%

Average price of the 30% most contacted listings

• Using the "Contacts CSV list", report the average price(format: € #,-) of the 30% most contacted listings.
• Output format is free for you to choose, but an example could be:

Average price
€ 12.512,-

The Top 5 most contacted listings per Month

• Using the "Contacts CSV list", report which listing had more contacts in each month.
• Reported fields: Ranking, Listing Id, Make, Selling Price (format: € #,-), Mileage(format: # KM), Total Amount of contacts
• Output format is free for you to choose, but an example could be:

Month: 01.2020

Ranking	Listing Id	Make	Selling Price	Mileage	Total Amount of contacts
1	1000	BWM	€ 2.538,-	50.000 KM	15
2	1001	Audi	€ 4.300,-	20.000 KM	12
3	1002	Toyota	€ 18.250,-	35.000 KM	11
4	1003	VW	€ 25.080,-	45.678 KM	10
5	1004	Porsche	€ 102.000,-	2.000 KM	8

Month: 02.2020

Ranking	Listing Id	Make	Selling Price	Mileage	Total Amount of contacts
1	1004	Porsche	€ 102.000,-	2.000 KM	18
2	1001	Audi	€ 4.300,-	20.000 KM	17
3	1000	BWM	€ 2.538,-	50.000 KM	15
4	1003	VW	€ 25.080,-	45.678 KM	10
5	1002	Toyota	€ 18.250,-	35.000 KM	3

Definition of the CSV files

• Listing.csv

field	type	required
id	numeric	yes
make	alphanumeric	yes
price	numeric	yes
mileage	numeric	yes
seller_type	alphanumeric	yes

• contacts.csv

field	type	required
listing_id	numeric	yes
contact_date	UTC Timestamp(ms)	yes

Setup and execution

To run the application you will need the following:

• Node.js installed on your machine
• A clone of this repository
• Inside the project directory run npm install in order to install all the dependencies.
• Start the server with npm start
• Navigate to http://localhost:3000 to get a review all the reports.
• Make a get request to API endpoint http://localhost:3000/api/reports to receive a json file with all the reports in a sctructured format.

Tests

At the present state of the repository, the only tests that are provided are for the function #avgListingSellingPrice. To run all tests simply execute the command npm test inside the project's directory.