Yan Zhang
Instacart Grocery Market Analysis
Tools:
Jupyter
Python
Project Context
Instacart is a leading online grocery platform seeking to better understand customer behavior to improve targeted marketing strategies. Stakeholders want insights into how different customer groups shop, when they shop, and what products drive the most sales.
Objectives and key questions:
The Rockbuster Stealth Management Board has asked a series of business questions and they expect data-driven answers that they can use for their 2020 company strategy. Here are the main questions they’d l ike to answer:
Identify the busiest days and hours to optimize ad scheduling during slower periods.
Discover which product categories or departments generate the highest order volume.
Determine when spending is highest to align product promotions with peak purchase times.
Understand customer segments and how their purchasing behaviors differ.
Simplify products into price range categories to support targeted marketing.
Data accessed from Kaggle.
Data cleaning
All missing values in the days_since_prior_order column occur when order_number = 1, confirming that these null entries represent customers’ first orders with no prior purchase history. Therefore, these records were flagged as "first_order" in a new column for clarity and further analysis.
Question: When are the busines hours of the day and days of the weeks?
The majority of orders were placed between 9:00 a.m. and 5:00 p.m., whereas order activity was minimal between midnight and 6:00 a.m.
Orders peaked on Saturday and Sunday, while Tuesday and Wednesday were the quietest days.
Question: What's the distribution among users in regard to their brand loyalty?
The chart shows that Regular customers make up the majority (55.1%) of the total customer base.New customers account for roughly a third (37.3%), while Loyal customers represent a small minority (7.6%).
Question: What particular time do people spend the most money?
Average spending reached their lowest point around 1:00 a.m. and peaked at approximately 3:00 a.m. Between 8:00 a.m. and 3:00 p.m., Average spending remained relatively stable.
On average, average spending peaked on Monday and reached their lowest point on Friday.
Question: What particular time do people spend the most money?
Produce, dairy eggs, snacks, beverages, frozen, and pantry were the most popular departments.
Question:What are the customer profiles by region?
The majority of customers were located in the South, followed by the West, Midwest, and Northeast. The distribution of customer profiles was relatively consistent across regions. Low- and medium-income parents were the most frequent shoppers, followed by parents with one child, while young adults without children shopped the least often.
Question: Are there any difference in ordering habits by customer profile?
South and Midwest customers demonstrated higher willingness to spend compared to those in the Northeast and West. New customers were more willing to shop and spend than loyal or regular customers.
Recommendations
Demand Optimization
-
Promote low-order days (Tues & Wed)
-
Run ads during 3–10 PM to increase engagement
Product Strategy
-
Prioritize stock for high-demand categories
-
Maintain inventory to avoid lost sales
Customer Retention & Growth
-
Introduce loyalty programs
-
Target high-spending new customers with personalized offers
Market Expansion
-
Focus on growth in underrepresented regions (Midwest & Northeast)
Segmentation Strategy
-
Value offers for budget-conscious families
-
Premium perks for high-income customers
Deliverables
