Case Study

Client:

Allbirds

​

Challenge:

The client needed to predict future customer lifetime value (CLV) to inform their marketing and retention strategies. The initial dataset contained incomplete information and negative values, requiring extensive cleaning and preparation to build an accurate predictive model.

​

Solution:

We utilized RStudio to build a linear regression model. The process involved:

• Data Cleaning and Preparation: We removed negative values and incomplete records from the dataset, creating a reliable foundation for our analysis. We also engineered new columns for sales, purchase counts, and purchase frequency to enrich the data for the model.

• Model Building: We focused our analysis on repeat customers who had made more than one purchase. We then split the data into an 80% training set and a 20% testing set to train and validate the model.

​

Results:

The linear regression model was a resounding success, providing actionable insights with a high degree of accuracy.

• Predictive Power: The model successfully identified key factors that predict CLV, enabling the client to forecast future customer revenue.

• Actionable Insights: The analysis revealed that the average time between purchases was under 50 days, and that most of the repeat customers made fewer than 10 purchases. This insight is crucial for developing a retention strategy.

​

Conclusion:

This project successfully established a highly accurate predictive model for customer lifetime value. The insights gained from this analysis are invaluable, providing a clear path forward for the client. The model can now be used as a powerful tool to identify and cultivate high-value customers, enabling personalized marketing campaigns and driving significant business growth and profitability. The next step is to integrate this model into the client's marketing platform to begin targeting these valuable segments.