Maximizing the Expected Revenue: The Use of Machine Learning Models for the Case of a Soccer Ball Company
Cargando...
Fecha
2022-12
Autores
García-Meléndez, Fátima A.
Título de la revista
ISSN de la revista
Título del volumen
Editor
ITESO
Resumen
Descripción
Every company begins with a fundamental questions and re-asks this questions several times throughout the company´s life, what is the product that will create enough value for it´s customers so that enough money can be charged to make a profit and keep on creating more value? So basically, a company´s strategy begins with a great product design and a price tag that customers are willing to pay that will maximize revenue. In past years, pricing has been so unattended because usually the responsibility tends to fall under different areas of the company and due to it´s complexity, task associated with setting prices are often not on top of the incumbency list. Ergo, prices are not varied enough for different product items, market segments and purchased occasion, impacting the demand, sales and perceived value of the product and brand. Therefore, this study aimed to determine how can machine learning models help create value and maximize revenue by determining the best product and price for a soccer ball company in Mexico. As a result, this research was able to determine that there are 3 different customer segments and that each of them values different characteristics of the soccer ball. Also, that a random forest model was the best model to calculate the purchase probabilities compared to a naive bayes model, a general linear model with logit link and a support vector machine model. Given those probabilities, the expected revenue was calculated for all the different product profiles, or combinations of the ball, and concluded that a price discriminated model with 3 balls; 1 targeted for each customer segment, can increase the expected revenue from an approximate of $166 to $1,572 dollars, proving that machine learning models and information-based decision making processes should be a must for every company
Palabras clave
Pricing, Maximization, Value, Product Design, Machine Learning, Random Forest, Expected Revenue
Citación
García-Meléndez, F. A. (2022). Maximizing the Expected Revenue: The Use of Machine Learning Models for the Case of a Soccer Ball Company. Trabajo de obtención de grado, Maestría en Ciencia de Datos. Tlaquepaque, Jalisco: ITESO.