Welcome to the E-Commerce Customer Analysis & Predictive Modeling project! 🚀 This project explores customer behavior through data visualization and predictive modeling, helping us understand what drives customer spending.
Unlike the app, time on the website does not strongly correlate with spending. This suggests improving the website may not significantly impact revenue, while the app experience holds greater potential.
We see a positive correlation between time spent on the app and yearly spending. Investing in the app experience could drive higher revenue!
A hexbin plot shows the density of data points. Customers who spend more time on the app are not necessarily long-term members, meaning app engagement doesn't directly correlate with loyalty.
A pairplot helps us visualize relationships between different numerical variables. From this, we can see which features have strong correlations with Yearly Amount Spent, hinting at key predictors.
One of the strongest correlations! Customers with longer memberships tend to spend more yearly. This insight suggests loyalty programs could be a major driver for revenue growth.
Here, we compare our model's predictions to the actual test values. A strong diagonal alignment confirms our model is performing well!
This histogram visualizes the residuals from our linear regression model, showing how well our model fits the data. A normal distribution indicates a good model fit, whereas skewness may suggest underlying patterns the model hasn't captured.
- Holding all other features fixed, a 1 unit increase in Avg. Session Length is associated with an increase of $25.98 in total dollars spent.
- Holding all other features fixed, a 1 unit increase in Time on App is associated with an increase of $38.59 in total dollars spent.
- Holding all other features fixed, a 1 unit increase in Time on Website is associated with an increase of just $0.19 in total dollars spent.
- Holding all other features fixed, a 1 unit increase in Length of Membership is associated with an increase of $61.27 in total dollars spent.
💡 Should the company focus more on the mobile app or the website?
This is a tricky question! There are two ways to think about it:
- Improve the website to catch up to the performance of the mobile app.
- Double down on the mobile app, since it already drives more revenue.
The best decision depends on additional factors, such as customer feedback and engagement metrics. It may be useful to explore the relationship between Length of Membership and Time on App vs. Time on Website before finalizing a strategy!
✅ Length of membership is the best predictor of yearly spending. ✅ Time on the app is more important than time on the website for increasing revenue. ✅ Our predictive model performs well, as shown by the residual histogram and test value scatterplot. ✅ Focusing on customer loyalty programs and app engagement could drive business growth.