Henry Ford famously said of the Model T, “you can have any color you want, as long as it’s black.” Today, in the digital age, the ability to personalize has become a key strategy for winning market share. Yet a century later, a ‘one-size-fits-all’ approach to supply chain has persisted well beyond its apparent sell-by date. For example, a survey of 100 organizations found that only 8% of manufacturers had achieved an advanced segmentation capability. How has this happened? Can supply chain segmentation underpin competitive advantage in the emerging age of micro-segmentation and a one-to-one paradigm?

What is supply chain segmentation?

The concept of supply chain segmentation is to match service characteristics to the needs of the customer at a profit. For example, customer needs may vary across expected service levels, promotional offers, delivery needs, and product customization. Also, product or SKU characteristics may determine supply chain configurations and inventory policies.

It doesn’t stop there: the implications have a profound effect on decisions made across the value chain. Impacts may range from stock coverage and positioning to sales forecasting, from consensual planning, arbitrations, and trade-offs to order promising, promotions, and emergency resolutions. In short, segmentation drives smart, profitable decision-making across the organization.

Why was segmentation underutilized?

Until recently, the full power of segmentation could not be delivered because business analytics were not up to the task. In manufacturing industries, segment populations can range up to 400,000 SKUs, while retailers can approach 10 million SKUs. Whatever business they are in, companies can quickly become overwhelmed. Therefore, segmentation maturity settled on a limited number of segments and simple parameters, such as ABC, Pareto models, or basic channels. The idea of creating more segments by mining the information in big data has been beyond the capacity of systems to ingest. It can take weeks to analyze a single segment with traditional methods and, as we know, today’s markets can be volatile and change rapidly. Moreover, it is tough for humans to consider the full complexity of parameters such as price, lead time, persona, variability, and a host of other factors.  

The Game-changer: Machine Learning

Fortunately, machine learning (ML) offers a time- and cost-effective solution for the complex mathematical problems underlying dynamic segmentation. Machine learning is adept and tireless at finding meaningful digital signatures in big data, along with cause and effect relationships. It can do the legwork of analyzing large volumes of data to help planners confidently and accurately answer critical questions, such as the attributes which should define segments, their weighting, the number of segments, and their definitions.

One company that has realized incredible value and achieved planning maturity is Mahindra & Mahindra, a $20 billion Indian automotive company, which reduced its inventory stock by up to 10% by coupling dynamic segmentation with an inventory optimization solution. They also realized much wider benefits, including improved accuracy with demand profiling and clustering, and more efficient allocation. There is now a systematic, strategic approach that tailors a unique supply chain for each demand cluster.

The Pandemic as a Catalyst

The pandemic has only served to reinforce the financial logic that underlies segmentation. Under conditions of limited funding or product scarcity, what is the best equation to optimize growth and generate value? The answer lies in defining differentiated service models, including pricing, that helps to increase market share and support growth. As markets become increasingly digital with further complexity and volatility, it is critical to examine the new generation of segmentation technology.

It is ironic to note that the new secret weapon of supply chain has been a long-standing capability historically underused. However, this new generation of supply chain segmentation fulfills the potential it always promised. At Blue Yonder, we call it «dynamic segmentation» and consider it a key capability in creating value-added and differentiated service at profit. Segmentation is re-emerging as a core driver of competitive advantage in the new age of supply chain.