One of America’s leading food manufacturers in an innovative and rapidly growing industry segment was collecting volumes of data but realized that the data being stored was not being utilized to drive further customer insights and enhanced decision-making through improved customer segmentation. Recognizing the need for both expertise in the industry and the analytics field, the client selected Clarkston Consulting to partner on this endeavor.
Wishing to capitalize on the available data for actionable business and customer insights, the client tasked Clarkston with identifying opportunities to capitalize on the rapidly growing customer demand with efficiency and scalability.
Clarkston’s data scientists were able to take sales data the company was already collecting, and then group customers based on buying patterns – all using an unsupervised machine learning algorithm. This new perspective of their customers showed gaps in the company’s current allocation process and provided insights with a clear action to make changes.
The primary objectives for the team was understanding the capabilities and data needed to create insights and opportunities to effectively grow with customer demand. This was a strong goal for the company. This project resulted in a developed hypothesis that focused on the exploration of insights in the data set, as well as utilization of a machine learning algorithm to cluster their customer base on sales revenue by product and by total revenue. This provided a deeper analysis and understanding on each cluster’s revenue over time, sales revenue by product over time, and the number of customers who placed an order in each cluster over time. The team in closing recommended potential opportunities to grow sales in certain products for specific customers.
Some of the key benefits from this project were the perspectives given on the client’s customer segmentation from the traditional approach of categorizing customers through geography and size using an unsupervised learning algorithm to detect the sales revenue patterns. This project overall shed light on the effects of the client’s allocation problem on each customer’s spend per product. This project proved the value opportunity for capitalizing on existing data outside of current efforts to improve customer segmentation and in turn increase sales.
Contributions by Elise Watson.