← All data exercises

Assess how a new delivery market in New York is going, and audit its data

1 table · 6K rowsUnlock with Pass

Jumpman23 is an on-demand delivery platform connecting “Jumpmen” (couriers) with customers who want a variety of goods. Jumpman23 sends Jumpmen to merchants to buy and pick up the items a customer requested; whenever possible, it orders the items ahead to save the Jumpman time. Each completed delivery is saved to the Jumpman23 database. The company is growing fast and has just launched in its newest market: New York City.

The CEO has just asked you “how are things going in New York?” He has also heard reports of data integrity issues. Think through what the CEO is really asking, analyze the market, and dig into the data integrity reports: if the issues exist, show where they are and how they affect the analysis. The CEO is a visual learner, the CTO loves to see code and technical work, and both are obsessed with maps.

The data

  • deliveries — a sample of completed New York deliveries, one row per item line: a delivery with several items spans several rows, so first decide what one delivery is.
    • delivery_id: unique id of a delivery. customer_id: the customer. jumpman_id: the Jumpman who completed the delivery. vehicle_type: how the Jumpman travelled to complete the delivery.
    • pickup_place: name of the pickup location. place_category: a categorization of the pickup location.
    • item_name, item_quantity: the item requested and how many. item_category_name: the merchant’s own categorization, such as “appetizers” or “soups”.
    • how_long_it_took_to_order: how long it took to place the order, as a duration.
    • pickup_lat, pickup_lon, dropoff_lat, dropoff_lon: coordinates of the pickup and dropoff locations.
    • when_the_delivery_started, when_the_Jumpman_arrived_at_pickup, when_the_Jumpman_left_pickup, when_the_Jumpman_arrived_at_dropoff: local timestamps of each stage of the delivery.

Your task

  1. How is New York going? Define the handful of metrics you would use to judge the health of a newly launched market (demand, customers, Jumpman supply, merchants, delivery speed), explain why those, and describe how they are trending over the period the data covers.
  2. Where is the business happening? Map pickups and dropoffs, and show where demand is concentrated, which merchants and categories it depends on, how far deliveries travel, and where the market looks underserved or over-reliant on a few merchants or customers.
  3. Where does the time go in a delivery? Break delivery time into its stages and compare it by vehicle type, place category, distance, time of day and whether the order seems to have been placed ahead. What would most improve the customer’s experience?
  4. Do the data integrity issues exist? List the problems you find, show where they occur, and for each one say how it affects the results above and what you did about it. What would you ask engineering to fix, and in what order?
  5. What should Jumpman23 do next in New York, and what data that isn’t here would you want before committing to it?

What to deliver

One presentation for the CEO and the CTO together: a visual main body, led by maps and charts, that answers the CEO’s question, plus a technical appendix for the CTO with the code behind each number and the details of the data integrity checks.

All company names in this question are fictional.

Data preview

deliveriesfirst 5 rows
delivery_idcustomer_idjumpman_idvehicle_typepickup_placeplace_categoryitem_nameitem_quantityitem_category_namehow_long_it_took_to_orderpickup_latpickup_londropoff_latdropoff_lonwhen_the_delivery_startedwhen_the_Jumpman_arrived_at_pickupwhen_the_Jumpman_left_pickupwhen_the_Jumpman_arrived_at_dropoff
1457973327168162381vanMelt ShopAmericanLemonade1Beverages00:19:58.58205240.7446-73.990740.7521-73.98542014-10-26 13:51:59.898924NULLNULL2014-10-26 14:52:06.313088
137705664452104533bicyclePrince Street PizzaPizzaNeapolitan Rice Balls3Munchables00:25:09.10709340.7231-73.994640.7197-73.99192014-10-16 21:58:58.6549102014-10-16 22:26:02.1209312014-10-16 22:48:23.0912532014-10-16 22:59:22.948873
147654783095132725bicycleBareburgerBurgerBare Sodas1Drinks00:06:44.54171740.7285-73.998440.7286-73.99512014-10-28 21:39:52.6543942014-10-28 21:37:18.7934052014-10-28 21:59:09.9848102014-10-28 22:04:40.634962
1485494271149157175bicycleJuice PressJuice BarOMG! My Favorite Juice!1Cold Pressed JuicesNULL40.7389-74.002740.7513-74.00562014-10-30 10:54:11.5318942014-10-30 11:04:17.7595772014-10-30 11:16:37.8958162014-10-30 11:32:38.090061
1327707122609118095bicycleBlue Ribbon SushiJapaneseSpicy Tuna & Tempura Flakes2Maki (Special Rolls)00:03:45.03541840.7261-74.002540.7093-74.01592014-10-10 00:07:18.4505052014-10-10 00:14:42.7022232014-10-10 00:25:19.4002942014-10-10 00:48:27.150595