How a Scenario-Based Pricing Tool Helped a Global Coffee Chain Strengthen Margin Decisions

Key Highlights: What This Case Study Covers

• Custom-built pricing tool and scenario planner for SKU-level margin optimization, revenue growth, and pricing decisions
• ML-driven price elasticity models for large-scale store-SKU combinations using Bayesian and OLS techniques
• Integrated SLS QP optimizer aligned with real-world business constraints, transaction thresholds, and category guardrails
• Interactive modules for scenario comparison, price simulation, and business impact evaluation

Client Overview

The client is a multinational coffeehouse chain with operations across 30+ countries and more than 2,500 stores. They were looking to move beyond manual pricing methods and adopt a scalable, data-driven approach that could support consistent pricing decisions across food and beverage categories.

The Ask

The client needed a transparent pricing engine that could recommend SKU-level price changes based on elasticity signals, simulate business impact under different scenarios, and streamline the execution of price updates.

Challenges

• Data access controls: Pricing data lived across multiple systems with varying access rules.
• High-volume scaling: The solution needed to support more than 200 SKUs across 2,500+ store combinations.
• Sparse data: Several SKUs had limited historical records, creating gaps in input coverage.
• Manual processes: The existing workflow depended on spreadsheets and manual judgment.
• Limited transparency: The previous pricing system offered minimal visibility into how prices were set.
• Model considerations: The pricing models had to balance accuracy with speed to ensure timely decisions.

Our Solution: A Unified Pricing and Scenario Planning Platform

Integrated Web Application

A standalone web application brought all pricing workflows into one interface, built to handle high-volume computation for 2,500+ stores and 200+ SKUs. It replaced manual spreadsheets with a consistent, model-driven process.

Price Recommender Module

Users set objectives, applied rule sheets and constraints, and excluded items when needed. The module triggered the optimizer to generate SKU-level price recommendations based on elasticity signals and business guardrails.

Scenario Planner Module

Users adjusted SKU-level prices to test what-if scenarios driven by promotions, competitor shifts, or business judgment. The SLS QP optimizer applied price bands, transaction thresholds, and other real-world constraints to keep scenarios actionable.

Final Recommendation

All scenarios were consolidated in one view, allowing comparison of revenue, margin, and volume changes in both percentage and absolute terms. Teams could quickly identify and lock the pricing set ready for rollout.

Price Performance Module

Teams tracked estimated vs. actual outcomes after implementation, with trend views for sales, gross margin, and transactions across monthly, quarterly, and annual periods.

Competitor Pricing Module

Item-level comparisons between the client’s stores and competitor outlets offered clear visibility into price differentials and supported informed price-positioning decisions.

Impact Delivered

• Provided SKU-level control over price changes while respecting business guardrails and elasticity thresholds.
• Enabled teams to simulate and compare multiple price sets with projected impact on margin, revenue, and unit sales.
• Increased transparency and traceability across pricing decisions.
• Supported live deployment within the client’s internal RGM environment.