
INSTACART
Analyzed data to find customer preferences and sales patterns by using Python 3 to clean, integrate, and visualize data provided by Instacart.
Data
Fictional customer base
https://d.docs.live.net/acfb05d6f4fa6932/Documents/Master%20folder/Project%20Management/Data
​
Analytical Techniques
-
Data Wrangling
-
Cleaning with Python
-
Data merging Deriving variables
-
Grouping and aggregating data
-
Manipulating data
-
Frames Data
-
visualization with matplotlib and Seabor


-
Exploring basic patterns such as busiest days, days with most orders or busiest hours of the day
-
Used Python commands to wrangle and clean data
-
Connected data sets
-
Performed commands to group, aggregate and summarize
-
Derived insights and used Python to create visualizations


-
Customer profiles were divided by age, loyalty, family size, and marital status.
-
They were further grouped and grouped by regions
-
This provided insights into customer’s spending habits based on different socioeconomic backgrounds

Results
What we can see is that the regular customers mostly consist of the middle-aged people who are buying lower priced items more frequently. They are order produce, milk eggs, snacks, beverages, and frozen items most often. We can also see a large number of our regular customers live in the Southern states.

​Recommendations
-
What we can do to increase other regions to have a higher regular customer population is to advertise more. For example, the West is close behind with regular customer numbers, then Midwest, and Northeastern regions.
-
Reflect and see if advertising is higher in the Southern regions than the Northeast. Advertising more or if it is a delivery issue, promote to delivers more.
-
To try to increase the Regular customer number we can take the Loyal customers and give them incentives to become Regular customers. Incentives could be anything from a rewards program where they purchase a certain dollar amount in a month, and they get a certain dollar amount off their next order.