Objective

As an analyst for Instacart, an existing online grocery store operating through an App, discover hidden information about its users and sales patterns. Perform an initial data and exploratory analysis of last year's data to derive insights and suggest strategies. Instacart stakeholders are seeking useful insights about their customers in their database in order to form a target marketing strategy. Use anything from customer profiles and purchasing behavior (by department, time of day, price) to present to stakeholders.

Key Questions

What are the busiest days of the week and hours of the day?
Are there particular times of the day when people spend the most money?
How to divvy up Instacart's products by price tags? Can we categorize the products by simpler price range groupings?
Are there certain types of products that are more popular than others (by Department)?

The Process

Consistency Checks

Import dataset into Python, and perform consistency checks to ensure data is ready for analysis (duplicates, null, and non-uniform data).

Exploratory Analysis

Now that data is validated, answer the Key Analysis Questions posed by stakeholders.

Stop & Strategize

Consider which information and data is worth presenting, and strategize how to best showcase these insights to stakeholders.

Create deliverables

According to the stakeholder questions, create helpful visualizations to clearly inform the Instacart stakeholders.

Challenges

Selecting how to categorize Instacart customers required strategic discernment. The database could categorize Instacart customers by age, marital status, income, customer loyalty, and many other factors, so it was difficult to determine which ones impacted purchasing habits the most.
- Resolution: Never assume your hypothesis is always correct. Test every hypothesis without bias, and trust the data's results (I initially thought age would be a bigger factor, but it turned out marital status played a much bigger role).
Python was a new language to me, and this project was my first opportunity to apply it to a real-life situation. Understanding how this fit with Jupyter Notebook was tough to navigate.
- Resolution: With lots of practice and exposure to Python and Jupyter, the more familiar I was with it. Also, asking for help and researching efficient ways to script code is always an option.

Visualizations

Recommendations

I recommend targeting ads when Instacart is attracting fewer orders,
or from 10 pm to 5 am.
Since Instacart sales skyrocket over the weekend, I recommend including promotions to purchase other department items during these times.
Marketing and advertising should be concentrated on the Middle-Class Parent population, as that is by far the highest Instacart Customer Profile.
If Instacart customers spend more expensive items in the morning, I recommend more advertisements be shown for these types of items around the same time.
Capitalize and create incentives for re-ordering common items in order to promote customer fidelity (Department ID's which are high: Produce and Dairy).

Skills used:

Python
Jupyter Notebook
Excel
GitHub