Optimizing Client-Specific Discounts to Maximize Gross Margin

In today’s highly competitive business landscape, where new companies and industries are emerging at an accelerated pace and global economic conditions remain uncertain, applying the right pricing strategy for each customer is more critical than ever. A well-structured pricing approach not only strengthens a company’s market position but also ensures sustainable profitability in the long term.

Pricing is adelicate trade-off between maximizing revenue and maintaining customer loyalty. If prices are set too high, there is a risk of losing clients to competitors. On the other hand, pricing too low can reduce profit margins and devalue the company’s offerings. Striking the right balance requires a deep understanding of market dynamics, cost structures, and customer behavior.

This challenge is even more pronounced in the metallurgical and steel industry, where material prices are heavily influenced by external factors such as fluctuations in raw material costs, supply chain disruptions, and economic policies. In this context, for the national leader, an effective pricing model is not just acompetitive advantage – it is a necessity.

We recognize the complexity of pricing in the metallurgical and steel sector, and our expertise,combined with a data-driven approach, enabled us to create a tailored pricing strategy for this company. The company is responsible for meeting the demand of more than 5,000 clients with a portfolio of products segmented across 12 different families. Our methodology involved developing a value-based pricing strategy, which included segmenting clients into clusters based on their shopper profiles, price sensitivity, and loyalty to the company. Lastly,we determined the optimal maximum discount to offer, maximizing gross margin while not drastically impacting the chance of clients accepting the proposed offer, thus maintaining a strong and sustainable leadership position in the industry.

The proposed solution was built around two main streams. The first focused on challenging the company to redefine its existing customer clusters by incorporating a more data-driven approach. This approach aimed at establishing well-defined groups composed of clients with similar attributes. The process involved analyzing critical variables such as customer context and demographics, purchasing behavior, historical transaction data – including volume and margin—and the risk associated with each client. As a result, customers were segmented into six distinct groups, each characterized by specific traits and purchasing patterns, allowing for a more precise and strategic discount allocation.

The second stream focused on developing an analytical elasticity model designed to predict, based on historical purchasing behavior, the likelihood of a customer accepting a given discount. This model utilizes machine learning techniques, where historical data on customer purchases, pricing, and discounts served as in put features. By training on this data, the model identifies patterns and relationships between customer characteristics, past purchase behavior, and their response to various pricing strategies. The model generates a set of price elasticity curves that consider order details and client attributes, providing a projection of the likelihood of offer acceptance. These curves were quite dynamic, adapting to different client segments and product families,allowing for a more precise and tailored approach to pricing decisions. Examples of the curves generated at the cluster and product levels are presented in the images below.

‍

‍

By analyzing past purchasing data, the model assesses how customers have responded to different discount levels. It identifies patterns in order volume changes in response to discount variations, helping to determine whether a client is highly sensitive to discounts or if their purchasing decisions remain relatively stable regardless of price reductions. As a result, a set of curves was generated which, once aligned with the business requirements and constraints, can be used to find the most suitable balance between discounts offered to different client segments.

After a comprehensive analysis, the tailored discount table proposed in our solution revealed a promising potential growth in gross margin by up to approximately 28% with a minimal impact on the acceptance rate and overall company revenue of less than 1%. This outcome highlights the effectiveness of the strategy in optimizing profitability without significantly compromising sales volume and customer satisfaction. By striking a careful balance between discount allocation and business constraints, the approach ensures that the company remains competitive while strengthening long-term financial sustainability.

This project was carried out through a series of meetings with the company's stakeholders to validate the results of each stream and fine-tune key aspects of the approach. Their insights and feedback were crucial in ensuring that the proposed solutions aligned with the company's strategic goals and operational realities. This collaborative process also facilitated a deeper understanding of potential challenges and allowed for adjustments to enhance the practical implementation of the strategy.