When your prices don't move, your elasticity models go blind.Here's how to see again.
EDLP is now the default pricing model for the world's largest retailers, such as Walmart, Home Depot, Aldi, and Lidl. But the analytics stack most marketers rely on was built for the opposite world: frequent promotions and big price swings. When prices sit flat for months, the standard elasticity toolkit quietly stops working.
This session introduces a demand-modeling framework designed specifically for stable-price environments and shows how it recovers reliable own- and cross-price elasticities, willingness-to-pay, and competitive substitution - using data most retailers already have.
Nearly every established pricing model, SCAN*PRO, log-log elasticity frameworks, promotion-decomposition methods, depends on one thing: prices that changes often (think of deep-price discounting). That assumption holds under High-Low (or commonly known as Hi-Lo) pricing. It collapses under Every Day Low Pricing (EDLP), where the whole point is price stability. The result is a growing blind spot for pricing and category teams at exactly the retailers and brands where EDLP now dominates.
This webinar walks through an alternative that doesn't rely on price moving over time. Instead of tracking a single SKU as its price changes, the approach organizes SKUs into demand groups of close substitutes and exploits the price differences across products competing for the same shopper. That cross-sectional variation is enough to estimate demand rigorously - using a Multiplicative Competitive Interaction (MCI) model for market share and a mixed log-log model for quantity.
For marketers, the payoff is concrete. You get SKU-level own- and cross-price elasticities in categories where you previously had none. You can quantify willingness-to-pay and map the "zones of indifference" where a price change won't shift share — critical intelligence when you only get to reset prices a few times a year. And because the method has been validated on both a durable category (water heaters) and a fast-moving one (soup), across both EDLP and Hi-Lo chains, it travels across very different parts of the assortment.
Topics covered
- Why standard elasticity models struggle when prices stay flat, as they do under EDLP
- Using price differences across competing SKUs, rather than price changes over time, to estimate demand
- The modeling approach: an MCI model for market share and a log-log model for quantity
- Estimating own- and cross-price elasticity at the SKU level
- Willingness-to-pay and zones of indifference, and why they matter when prices reset infrequently
Key takeaways
- When a stable-price environment may be affecting your current elasticity estimates
- How to group SKUs so that price differences between them become usable data
- How the MCI and log-log models produce elasticity estimates without frequent price changes
- How to read elasticity, willingness-to-pay, and zones of indifference in a pricing decision
Recordings of AMI webinars are available on demand exclusively to AMI members via the AMI Member Hub. Log in using your member credentials through the AMI website main page, then navigate to the Professional Development section to access the webinar recordings.
About the presenter:
Dr. David Sugianto Lie
Senior Lecturer of Marketing, Monash University
Dr David Sugianto Lie is a Senior Lecturer in Marketing and an applied modeling specialist who works at the intersection of marketing theory and commercial practice. His core focus is building statistical models that help companies make sharper decisions, spanning retail and pricing strategy, segmentation, optimisation, and customer experience, drawing on both traditional econometrics and machine-learning methods.
That work has translated directly into industry. David has consulted for a range of organisations, including Fortune 500 company Coca-Cola Amatil and Telecom Malaysia, helping them turn data into practical pricing and marketing decisions. He holds a PhD in Marketing from the University of New South Wales and has taught across the Marketing and Applied Econometrics disciplines at both undergraduate and postgraduate levels for eight years.
His research carries the same practical bent. He holds a grant from the Marketing Science Institute (MSI) and Australian Marketing Institute (AMI), and his work has been published in leading journals including the Journal of the Academy of Marketing Science, Industrial Marketing Management, the European Journal of Marketing, and Decision Support Systems. He has presented at the Marketing Science Conference, ANZMAC, and the American Marketing Association.
