Using Machine Learning for Price Optimization

Operating in a highly competitive market, the global diagnostics company faced challenges in maintaining profitability and revenue growth. This led to Phronesis collaborating with the company to enhance its pricing strategy using machine learning (ML).

Objective

The diagnostics company aimed to address increasing competitive pricing pressures by developing a strategy that would:

• Drive profitability and revenue growth
• Consider the unique requirements of the diagnostics industry
• Implement dynamic pricing adjustments across various lab locations, product categories, and SKU families

The primary goal was to achieve daily price optimization tailored to specific market conditions and organizational objectives.

Solution

Phronesis Partners adopted a comprehensive approach to meet the client's objectives:

• Data Integration: Consolidated weekly sales data, product and sales hierarchies, competitor pricing information, and relevant SKU metadata into the company's existing SAP database, ensuring seamless data availability for analysis
• Machine Learning Modeling and Optimization: Developed ML models using Python and Alteryx to capture demand-price responses and predict sales volume and gross profit for each SKU and price point. Incorporated pricing rules and guardrails defined at enterprise, category, member, and location levels to ensure compliance with organizational guidelines. Balanced revenue and profit goals by considering member objectives in selecting the optimal pricing strategy
• Dynamic Pricing Execution:Implemented a daily optimization process that recalculated optimal prices based on defined rules and member objectives. Shared optimized prices with the company's pricing team for evaluation and approval. Utilized Power BI to visualize high-level summaries of revenue, profitability, opportunity curves, and the cost of rules for effective decision-making

Impact

Within the first three months of deployment, the diagnostics company achieved:

• Daily price optimization across all lab locations, product categories, and SKU families
• Successful attainment or maintenance of revenue and gross profit targets at 80% of its locations
• Enhanced pricing competitiveness within the market
• Improved overall profitability and revenue generation

This collaboration underscores the effectiveness of machine learning in optimizing pricing strategies, enabling businesses to respond dynamically to market conditions and achieve strategic objectives.