Size & Pack Optimization
Eliminate broken sizes and reduce handling costs with demand-true size intelligence
Size and pack decisions have an outsized impact on retail performance, yet they are often driven by incomplete data and static rules. When popular sizes sell out early while fringe sizes accumulate, retailers lose sales, increase markdown exposure, and frustrate customers. At the same time, inefficient pack configurations create unnecessary handling costs across distribution centers and stores, slowing inventory flow and increasing labor requirements.
o9 Size & Pack Optimization addresses these challenges by transforming size planning from a static, rule-based exercise into a demand-driven, data-science-backed capability. By analyzing granular sales patterns and correcting for stock-outs, the solution generates size curves that reflect true customer demand rather than constrained sales history. These insights are then translated into optimized pack configurations that balance store needs, visual requirements, and supply chain efficiency.
This approach allows retailers to simultaneously improve sell-through and reduce operational friction. Stores receive the right size mix for their local customer base, while distribution and store operations benefit from fewer manual adjustments and more efficient handling.
The cost of broken sizes and inefficient packs
Broken size runs are one of the most persistent sources of lost revenue in apparel, footwear, and sporting goods retail. When core sizes sell out prematurely, customers leave without purchasing—even if inventory technically exists. Meanwhile, excess fringe sizes accumulate, tying up working capital and driving markdowns.
The challenge is compounded by regional variation. Size demand differs significantly by geography, store format, and customer demographics. A size curve that works well in one market may be completely wrong in another. Yet many retailers continue to apply uniform size profiles across large store groups, masking true demand patterns.
Pack inefficiency further amplifies the problem. Pre-packs that do not align with store-level demand require additional handling at DCs and stores, increasing labor costs and delaying floor availability. Without a system that connects size demand to pack logic, these inefficiencies persist season after season.
From static size rules to demand-driven optimization
Leading retailers are moving away from static size curves and fixed pack rules toward demand-driven size intelligence. Instead of relying solely on historical sales, they correct for stock-outs, identify localized demand patterns, and continuously update size profiles as customer behavior evolves.
o9 enables this shift by linking size demand modeling directly to assortment, buying, and allocation decisions. Size and pack optimization becomes a living capability—continuously refined, operationally feasible, and financially grounded.
What Makes o9 Different
Demand-True Size Modeling
o9 corrects historical sales data for stock-outs, ensuring size curves reflect unconstrained demand rather than availability-driven distortion.
Pack Optimization with Operational Awareness
Pack configurations are optimized with full awareness of handling costs, MDQs, and vendor constraints—reducing downstream inefficiencies.
Localized Size Intelligence
Size profiles are generated at store or store-cluster level, capturing regional and demographic differences without over-fragmenting planning.
Direct Integration with Buying
Size and pack recommendations flow directly into buy plans and purchase orders, eliminating manual translation.
Industries Supported
Powered by the o9 Digital Brain
Size & Pack Optimization runs on the o9 Digital Brain, leveraging the Enterprise Knowledge Graph to connect product attributes, size demand, store characteristics, and operational constraints. This enables size intelligence to scale across millions of SKU-location combinations without performance degradation.
The cloud-native platform supports rapid recalculation of size curves and pack logic as demand patterns change, ensuring recommendations remain current throughout the planning cycle.
Modular by design, enterprise by default
The o9 Size & Pack Optimization solution is built on a high-performance, attribute-driven data model that translates localized demand into executable size and pack directives. Rather than treating sizing as a static input, the architecture continuously refines size intelligence based on real sales behavior and execution outcomes.
Core Building Block
Size Profiling
This component generates size curves based on historical sales data aggregated at the appropriate level, such as style, class, store, or store cluster. Size profiles are calculated using unit sales patterns and normalized to represent proportional demand across sizes, providing a foundational view of how customers purchase each product. These profiles serve as the baseline for both assortment planning and buy execution.
Localized Size Curve Management
Size profiles can be differentiated by store or store cluster to reflect regional, demographic, and format-driven differences in size demand. This enables retailers to avoid applying a single national size curve across all locations, improving size availability at the store level while keeping planning complexity manageable.
Pack Configuration Management
This capability defines and manages different pack structures, including pre-packs, bulk packs (singles), and hybrid configurations. Pack definitions include constraints such as maximum units per pack, allowed size combinations, and vendor requirements, ensuring pack structures are operationally feasible.
Buy Plan and Procurement Integration
Optimized size and pack outputs are fed directly into buy plans and purchase order generation. This ensures that procurement decisions align with localized demand signals and eliminates manual translation between planning and buying systems.
Advanced Building Blocks
Stock-Out Rectification
Advanced algorithms adjust historical sales data to account for lost sales caused by stock-outs. This ensures that size curves reflect true customer demand rather than constrained availability, significantly improving accuracy for core and fringe sizes.
Dynamic Size Clustering
Stores are clustered specifically based on size demand behavior rather than overall sales volume. This allows planners to distinguish, for example, stores that skew toward smaller sizes versus those with higher demand for extended sizes, improving allocation precision.
Pack Optimization Solver
A constraint-based optimization engine determines the optimal mix of pre-packs and singles by balancing demand fulfillment, handling costs, Minimum Display Quantities, and vendor constraints. This reduces DC and store labor while improving inventory efficiency.
Attribute-Based Size Modeling for New Products
For new products without sales history, size curves are generated using attribute similarity to historical items, such as brand, fit, silhouette, or price tier. This allows accurate size planning from the first buy.
A digital operating model for VUCA conditions
APEX is o9’s AI-powered operating model for enterprises navigating volatility, uncertainty, complexity, and ambiguity (VUCA). It enables organizations to plan, execute, and learn as one connected system.
The o9 Digital Brain powers APEX by connecting enterprise data, knowledge, and decisions through a single intelligent model.
Collaborative Demand Planning is one of the building blocks of the Digital Brain. It contributes domain-specific capabilities into the enterprise-wide model that enables APEX from the ground up—linking this solution to decisions across the entire value chain.
→ Learn how the APEX Operating Model works
Where AI drives real decisions
AI enhances Size & Pack Optimization by converting fragmented sales history into demand-true size intelligence.
Machine learning models correct historical sales for stock-outs and identify localized size demand patterns across stores and clusters.
Prescriptive optimization algorithms determine optimal pack configurations by balancing demand fulfillment, handling costs, and operational constraints.
Generative AI enables planners to query size and pack logic using natural language, while agentic AI automates curve refreshes, pack recalculation, and exception monitoring.
→ Learn more about o9 AI innovations
What our customers say
"We made the conscious decision with o9 to bring a quicker ROI by integrating with our legacy SAP. [...] When the full ERP transformation happens, we’re ahead of the game."
Paul Tips
Product Owner at Canyon Bicycles
"What's really succeeding with us is the idea of the connection to the data and a best-in-class UX/UI, so the people that use the business can really make an impact."
David Almeida
Chief Strategy & Technology Officer at Anheuser-Busch InBev
"With o9 AI/ML-based forecasting in place, we’re already seeing improved forecast accuracy, stronger cross-functional collaboration, and faster, more informed decision-making—all within a centralized platform."
Gaby Gutierrez
VP of Global Supply Chain Planning at Amway
Related solutions
Frequently Asked Questions (FAQ)
It is the process of aligning size demand and pack configurations with localized customer behavior and operational constraints.