Project Title:

Pawformance Boost

Project Description:

Pawformance Boost is a predictive analytics project aimed at enhancing the effectiveness of dog food advertising strategies to boost market share. Our team analyzed customer behavior and ad impact and developed a model to identify potential customers likely to purchase dog food post-ad exposure, enhancing targeting and ad efficiency.

Key Contributions:

●     Designed and implemented predictive models using Logistic Regression, SVM, Random Forest, and KNN, achieving up to 70% accuracy in predicting customer purchasing behavior.

●     Conducted extensive feature selection using univariate methods to pinpoint key determinants of ad effectiveness.

●     Developed the WOOF scoring system to prioritize high-potential customers for targeted advertising campaigns.

Impact:

●     Facilitated a five-fold increase in ad exposure for the client, significantly improving the precision of customer targeting.

●     Enabled the client to realign their marketing strategy, focusing on high-impact customer segments, potentially increasing market share.

Technologies Employed:

●     Programming Language: Python

●     Predictive Modeling: Logistic Regression, Support Vector Machine, Random Forest, K-nearest Neighbors

●     Feature Selection: Univariate Selection, SelectKBest

●     Data Handling and Analysis: Pandas, NumPy

●     Data Visualization: Tableau