Instacart Analysis of Sales and Customer Behavior

Project Overview

• Motivation: This analysis attempts to uncover sales patterns for an online grocery store, Instacart, to boost revenue for the company.
• Agenda: Understand general sales and customer behavior
  • What are the busiest hours and days of the week?
  • Are there certain products that are more popular than others?
  • What is the distribution among customers in terms of their brand loyalty?
  • Are there differences in ordering habits based on a customer’s region?
  • Is there a connection between age and family status in terms of ordering habits?

Data Details

• Datasets: Customers, Orders, Products, Departments
• Sources:
  • CareerFoundry (Customers)
  • Kaggle (Orders, Products, Departments)
    • “The Instacart Online Grocery Shopping Dataset 2017”, Accessed from http://www.instacart.com/datasets/grocery-shopping-2017 via Kaggle
• Tools, skills and methodologies:
  • Python was used for data cleaning and manipulation, as well as EDA, main analysis and creation of profiles, and data visualization.
  • Libraries used included pandas, NumPy, matplotlib and seaborn.
  • Provided a comprehensive presentation and final deliverable report in Excel.
  • GitHub repository for scripts and other analytical details can be found here.

Exploratory Data Analysis

• Conducted a simple time analysis to discern the most common time of day for online orders.
  • With 9 am to 4 pm showing the highest volume, my recommendation was to schedule more ads during this time.
• Another time analysis by day of the week showed which days were the busiest for online orders.
  • With Saturday and Sunday being the most common, my recommendation was to schedule ads for these days, since more people would see them, and therefore improve their efficacy.

Customer Profiling by Demographic

• Created various profiles using Python to discern customer behavior by demographic.
• These profiles included:
  • Family status
  • Age group
  • Level of income
  • US region
  • And more

Conclusion and Takeaways

• Through the different customer profiling and column derivations, I was able to uncover some key aspects that tended to increase customer behavior.
• Additionally, I garnered insights that allowed Instacart to improve its targeted marketing campaign. Some of these attributes included:
  • The most popular US region by number of customers.
  • The largest age group by usage.
  • Whether most customers are deemed ‘regular’, ‘loyal’, or ‘new’.
  • If you want the whole story, please refer to the full presentation and final deliverable, provided below.