On Webinar

Rolling Ball

About the webinar

Rolling Ball is an algorithm for FMCG product segmentation, it works by evaluating billions of rows of SKU level transaction data to find associations between products.  We can then use the dynamic segment as a base to weight customers based on their frequency of purchase in the segment.

On the surface this is a simple premise but the technical implementation to make this into a sustainable model is quite complex and the CX has made the outcomes very simple to consume and use. It reveals patterns in-between the products that are impossible for a human to spot or in many cases even hypothesise. For example people who buy organic food with a strong preference towards a certain kind of shampoo that would not readily be identified as organic. Thus we might get unexpectedly good results by targeting organic buyers in a shampoo promotion.

Have a SAS profile? To complete this form automatically Sign In


All personal information will be handled in accordance with the SAS Privacy Statement.

  Yes, I would like to receive occasional emails from SAS Institute Inc. and its affiliates about SAS products and services. I understand that I can withdraw my consent at any time by clicking the opt-out link in the emails.

About the Presenter

Lin Ding, Loyalty NZ

Lin is an analyst at Loyalty NZ Lab360 who has a passion for delivering value from data. He has been working with data for over 10 years; first  pursuing the truth in the world of statistics for 5 years, and then building solutions and answering questions with data and technology in a business context. His current role at Lab360 puts him on top of the billions of rows of FMCG transaction data and his focus is turning this into business value. He has built probabilistic models to characterize customer repeat transaction patterns, worked for Auckland Transport to analyze their on-street and off-street parking pricing strategy and uncovered the secret sauce hidden in the email subject line that drives people’s interest.