Reports
Optimising retail pricing under substitution effect
Your pricing model has a blind spot
When you discount one product, customers do not respond to it in isolation. They weigh it against every similar item in your range, and discounting one shifts demand across all of them. Most retailers do not lose profit because their prices are wrong. They lose it because their own products are quietly taking sales from each other.
This is product substitution, and most retail pricing strategies ignore it entirely.
The profit gap is real
Research shows that accounting for substitution in pricing decisions can increase retailer profit by up to 43.7% compared to standard approaches. That gap exists in every promotional cycle, including yours.
The reason it goes unaddressed is computational. A range of 100 products with 10 price points each generates more combinations than any system can evaluate directly. This report presents a framework that makes it tractable.
What you will get in the full report
• A clear explanation of substitution and cannibalisation, and why they make standard pricing approaches insufficient.
• A technical framework that incorporates pre-estimated substitution effects into pricing decisions across large assortments, formulated as a mixed-integer programming problem.
• A practical approach to managing the computational challenge, including how to classify SKUs by impact, limit substitute relationships intelligently, and reduce the search space without losing accuracy.
• A real-world application to a major fashion retailer, demonstrating how a problem that would take millions of years to solve naively can be made computationally feasible in under an hour.
• Guidance on how to incorporate business constraints and secondary KPI targets alongside the core profit objective.
This report is written for pricing, merchandising and data science leaders in retail, particularly those working in fashion, grocery, or any sector with high product similarity and frequent promotional activity.
If your team makes markdown decisions at scale, this gives you a rigorous framework to do it better.
Your pricing model has a blind spot
When you discount one product, customers do not respond to it in isolation. They weigh it against every similar item in your range, and discounting one shifts demand across all of them. Most retailers do not lose profit because their prices are wrong. They lose it because their own products are quietly taking sales from each other.
This is product substitution, and most retail pricing strategies ignore it entirely.
The profit gap is real
Research shows that accounting for substitution in pricing decisions can increase retailer profit by up to 43.7% compared to standard approaches. That gap exists in every promotional cycle, including yours.
The reason it goes unaddressed is computational. A range of 100 products with 10 price points each generates more combinations than any system can evaluate directly. This report presents a framework that makes it tractable.
What you will get in the full report
• A clear explanation of substitution and cannibalisation, and why they make standard pricing approaches insufficient.
• A technical framework that incorporates pre-estimated substitution effects into pricing decisions across large assortments, formulated as a mixed-integer programming problem.
• A practical approach to managing the computational challenge, including how to classify SKUs by impact, limit substitute relationships intelligently, and reduce the search space without losing accuracy.
• A real-world application to a major fashion retailer, demonstrating how a problem that would take millions of years to solve naively can be made computationally feasible in under an hour.
• Guidance on how to incorporate business constraints and secondary KPI targets alongside the core profit objective.
This report is written for pricing, merchandising and data science leaders in retail, particularly those working in fashion, grocery, or any sector with high product similarity and frequent promotional activity.
If your team makes markdown decisions at scale, this gives you a rigorous framework to do it better.
Get the full report
Access the optimisation framework that accounts for substitution across your entire range and start making markdown decisions that protect your margins.
DownloadTalk to our experts
Let's create real impact together with data and AI
Senior Manager, Retail
Petr Pushkar
Talk to our experts
Let's create real impact together with data and AI
Senior Manager, Retail
Petr Pushkar