Rotor credit failures in the Vetscan VS2, a veterinary diagnostic device made by Zoetis, impact the customer experience and cost Zoetis, as well as the customer, both time and money to manage. This paper presents a model for predicting rotor credit failures using customer, FUSE, and sales data. Three machine learning models were trained and evaluated: random forests, gradient-boosted decision trees, and multilayer perceptron classifier. Using factors such as AUROC and training time, a gradient-boosted decision tree was the best-fit model. Key variables influencing the model, including the lot number, total rotors run in 2023, and metrics for customer contacts in 2023, were identified. Using these variables, implications for business strategy, including improving data collection, better utilization of the available data, and further investigation recommended, were determined. Overall, this paper contributes to a deeper understanding of rotor credit failure and gives actionable insights for improvement and improving overall operation efficiency.