Instacart Project

Instacart is a grocery / household goods delivery and pickup service (available in US and Canada)
Customers order online (website / app) and have a personal shopper pick up at the stores and deliver on the same day
Offer 24/7 delivery service
Major retailers / grocers: Costco / CVS / Target

Motivation

Understand customer order behavior & repeat purchases
- help Instacart find new retailers or grocers to satisfy customer needs
Understand the traffic by day (Sunday to Saturday) and time (24 hours timeframe)
- help Instacart effectively allocate the number of workers to meet customer demand

Tools: DbVisualizer & PostgreSQL
Dataset provided by Modern Data Lab
- 248,081 rows
- 13 columns (Entity Relationship Diagram)
- data type: numeric & varchar
Tableau (link)

Custumer Order Behavior
- Customers commonly ordered 5 products in each purchase
- The top 10 products customers order are all fruits (Banana: 1st place)
- Fresh fruits and vegetables are also most commonly reordered
- On average, customers are 60% likely to reorder the same products
Instacart traffic
- Weekends have more customers and purchase an average of 2 more products in each order compared to weekdays
- Customers tend to shop in afternoons (2pm - 4pm) and in mornings (9am - 11am)
- The department (produce) and aisles (fresh fruits & fresh vegetables) are the busiest everyday in the afternoon

Partner with local fruit and veggie stores to meet customers' huge demand in fresh fruits and vegetables to prevent pontential shortage
Ensure a variety of products are in stock in different places as customers tend to make repeat purchases
Have more personal shoppers available on weekends and in 9 - 11 am and 2 - 4 pm to meet customer demand and satisfaction