Project Overview
Background
An online grocery store that connects with customers through a mobile app aims to refine its marketing strategy. By deepening their understanding into the different kinds of customers they have and how their order behaviour differ, they aim to leverage data driven insights to inform a more targeted marketing strategy.
Scope
The analysis examines data from the 2017 calendar year within the United States.
Goals
Perform an exploratory analysis of customer demographics and transactional data to gain insights into ordering behaviours, thereby enhancing customer profiling and optimising marketing strategies through targeted product placement.
Objectives
- Ordering Trends
Analyse order volumes, spending patterns, and product popularity across different price ranges to optimise advertisement scheduling and strategic product placement.
- Customer Profiling & Ordering Habits
Profile customers based on ordering behaviour and demographic information and analyse how order habits vary across these profiles.
Key Questions
- How does order volume fluctuate over time?
- How does average spending per product vary over time?
- How can products be categorised into price ranges to support product placement?
- Which products, within their respective departments and categories, are most popular?
- How can ordering behaviour and demographic information be used to categorise customers?
- How does ordering habits vary across customer profiles?
Tools & Techniques
- Python – Data Preparation | Exploratory Data Analysis | Visualisation
- Excel – Reporting
Data
This analysis uses publicly available data originally sourced from Instacart via Kaggle. The links as well as an additional customers dataset was provided by CareerFoundry as part of their Data Analytics Course.
- Customers – Customer ID, Name, Surname, Gender, State, Age, Date Joined, Dependants, Family Status, and Income
- Dataset
– Departments – Department id and name
– OrdersProducts – Order id, product id, add to cart order, and reorder indicator.
– Orders – Order is, order number, order day of week, order hour of day, days since prior order.
David Griesel 