AI can integrate supply and demand in manufacturing

“It starts with good customer segmentation,” Eaton said. This segmentation needs to align customers to the company’s growth strategy, customer revenue, customer margin, and other factors to help identify and prioritize customer segments appropriately.

“Doing that successfully requires a very, very precise understanding of cost to serve at a customer and a product level,” Eaton said. “If you don’t know the gross margin return on investment for serving that customer a given product or service, it’s very hard to make pricing decisions. It’s very hard to make investment decisions. It's very hard to even know how to calibrate and define the supply chain.”

Every company needs to identify which customers are core, and integral to its growth strategy. Then, companies can use their customer prioritization as fuel for supply chain decisions that help generate demand and reinforce the company’s differentiation in the market. “In other words, look at how you are creating a sustainable growth pattern, and sustaining the differentiation in the way that you serve customers,” Eaton said.

A dynamic connection between customer profiles and supply chain participants can help a company solidify its strategy for differentiation and growth. Many manufacturers are searching for that strategy — and they have the customer and supply chain data to help define it, but they just haven’t made the connection or realized that it can be done.

“Many of our clients are in a never-ending search for value and differentiation,” Eaton said. “That’s where we have an absolutely incredible opportunity to use artificial intelligence.”

Eaton described how an AI agent could combine data from a company’s CRM and SCM systems to ultimately inform maximum customer satisfaction and profitability. The agent could pull customer segmentation as well as data across a range of fields. “Then, look intrinsically at the operation of the business to determine how you are doing against your standard cost. Every company goes through a standard costing process at least yearly. Variances should be tracked in real time for proactive remediation of the root causes associated with negative variances.”

“An AI agent — if it has access to the right data — can identify variances and deliver the critical root-cause analysis that leaders need in order to drive decisions,” said Grant Thornton Technology Modernization Partner Adam Wengert. That analysis can include the customers served, the suppliers to be used, money made per customer and the operational execution and performance against standard cost. 

Show image description -->

“You’re picking those multiple dimensions of data, and then you are refining operational execution and the way that you strategically engage and sell to customers, to drive maximum customer satisfaction and maximum profitability for the organization,” Eaton said. “That’s the goal, and it is accomplished through the use of AI to bring together the necessary data and enable the analysis that would take hours in the absence of a robust data model and AI agent.”

“The importance of data in this approach cannot be overstated. This capability entirely depends on data, and a willingness to invest in AI,” Wengert said. Many companies already have both, but they just haven’t applied the use cases yet.

“Some companies report not realizing adequate returns on their investment in AI,” Eaton said. “Some of the early adopters are solving bits and pieces of this, but they’re not solving this problem fully. They might not understand the use cases. If they can understand these use cases, then they can make the right investment.” Use cases for this connected solution include:

• Tailored pricing and promotions: Link customer lifetime value and buying behavior with supply chain cost data to form dynamic pricing strategies or targeted promotions that protect margins while meeting demand, building customer loyalty and differentiating service.
• Intelligent pricing adjustments: Analyze raw material cost shifts against customer purchases, segmentation, prioritization and other factors to suggest adaptive pricing strategies that protect margins while maintaining relationships.
• Priority segment–driven production: Prioritize production for high-value customer segments or SKUs based on gross margin, cost to serve, historical buying patterns and other factors.
• Dynamic demand forecasts and inventory optimization: Forecast demand at the SKU level to adjust procurement for external events or priority customers.
• Supplier and procurement optimization: Evaluate supplier performance, pricing and other factors against customer demand forecasts to identify when to issue new RFQs, switch vendors or adjust order quantities to minimize risks and cost spikes.
• Risk management and disruption response: Detect early signals of supply chain disruptions to adjust plans and protect service levels for top-tier customers, using scenario planning and contingency development.
• Intelligent order fulfillment and routing: Monitor real-time logistics and inventory, rerouting shipments or adjusting warehouse slotting to minimize delays for critical accounts.
• Timely customer communication: Track supply chain status to give customers timely updates on accurate delivery timelines, reduce uncertainty, improve trust and help sales teams manage expectations.
• Sustainability reporting: Sourcing and logistics data can be integrated with customer purchase histories to give customers sustainability metrics that support ESG goals and reporting while differentiating value and strengthening customer relationships.

Schindler said enterprise-level AI capabilities are like the fascia tissue that connects, nourishes and contains muscles and organs throughout the body. Likewise, AI can facilitate the exchange and coordination of information across various systems to help the enterprise move and react as one.

This ability for an enterprise to move as one body becomes essential when leaders confront enterprise-scale questions.

Schindler noted that individual teams might address issues like supply chain risk and customer strategy, but leaders need enterprise-level integration to answer the bigger questions. “Do you know what your North Star is?” Schindler asked. “Where are you looking to go, and how are you getting there?”