Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Will AI disrupt traditional sales and operations planning (S&OP)?

It seems clear that AI will have a significant role in the future of supply chain management, but what’s the line between hype and reality? 

Will AI disrupt traditional sales and operations planning (S&OP)?

[Adobe Stock / AI Farm]

Profile photo of James Barlow

President · ORI

Just as the hype around artificial intelligence (AI) seemed unstoppable, Microsoft CEO Satya Nadella threw a massive wrench in things. On a podcast in late February, Nadella, whose company has poured billions into generative AI projects through OpenAI, said people need to get past the fantastical aspects of AI and demand to see real-world impacts. 

“The real benchmark is: The world growing at 10 percent. Suddenly, productivity goes up and the economy is growing at a faster rate. When that happens, we’ll be fine as an industry,” Nadella said on Dwarkesh Patel’s podcast. 

For product manufacturers and companies managing complex supply chains, Nadella’s words are especially relevant. Like other industries, supply chain software vendors have caught AI fever, embedding it into their products and frothy marketing campaigns. It seems clear that AI will have a significant role in the future of supply chain management, but what’s the line between hype and reality? 

Let’s focus on one of the most crucial aspects of supply chain management for most large and mid-size companies: alignment of supply, demand, and finance planning. This process, typically called “sales and operations planning (S&OP),” is where profits are made or lost, so it’s reasonable to ask, “Can AI make effective S&OP automatic?” 

THE ROLE OF SALES AND OPERATIONS PLANNING IN A VOLATILE WORLD

Before digging into how AI might (or might not) disrupt the S&OP process, let’s zoom out and examine why S&OP is so important. For companies that sell manufactured goods, supply and demand planning has arguably the biggest impact on their financial performance. A miss on forecasting can lead to excess inventory from being too aggressive or lost sales from being too conservative. Supply and demand planning is hard enough in a stable environment where global trade, currencies, shipping logistics, and other considerations are relatively innocuous. 

Unfortunately, the world is increasingly unstable as it relates to domestic and global supply chains. Software research analyst Robert Kugel wrote in early 2024 that supply chain managers must “be able to forecast and plan more nimbly, react to changes faster while making better informed decisions about tradeoffs” and “supply chains designed for the lowest possible cost [must be] restructured for resiliency in the current unsettled environment.” 

In other words, supply chains must account for managing uncertainty and optimize for resilience, which currently requires a highly effective human-led S&OP process, optimized with AI.

USING AI TO SUPERCHARGE S&OP

Most large and mid-sized product organizations recognize the critical place of S&OP in their business models, so it’s rational for those same organizations to look for ways that AI can replace human efforts by increasing efficiency and accuracy in this critical process. In fact, if AI can produce accurate forecasts, it’s reasonable to ask, why is S&OP needed at all?

There may come a day when S&OP can be fully automated, but we’re not there yet. Currently, a prudent approach is to look for ways AI can break down information silos and add operational efficiency and data accuracy to an S&OP process that still necessarily involves human intelligence. Logility notes that AI’s primary impact on S&OP is to remove the bottlenecks of manual data entry and communication, enabling instead the automated extraction and communication of real-time data insights across S&OP stakeholders. 

“GenAI and large language models have expanded S&OP capabilities. Not just to navigate around and quickly find things we’re looking for, but to assist us in analyzing information, providing insight, and making better decisions happen,” said Logility business consultant Joe Slater. 

In keeping with the theme of AI streamlining the efficiency and accuracy of a fundamentally human S&OP process, Deloitte, in late 2023, outlined the ways AI can “best add value and where to apply human intelligence.” As identified by Deloitte’s analysts, there are three primary use cases for AI in S&OP over the short to medium term: 

1. Forecasting Code Gen

Large language models (LLMs) can lower the technical bar for people to engage with forecasting models and give demand planners another tool to increase the accuracy of their forecasts with external data sets. Ostensibly, a demand planner could upload some historical sales data and a generative AI system could produce a forecast that accounts for variables like seasonality, externalities, and various forecasting methods. 

2. Structuring Unstructured Data

As any supply or demand planner knows, forecasts and plans are only as accurate as the underlying data. Some new GenAI use cases, such as converting product reviews into LLM data points to predict review scores, could be applied to turning unstructured data into structured data for forecasting. This use case is still theoretical. 

3. Assisting The Planning Team

The most immediate use case for AI in S&OP is organizing and summarizing large batches of data to streamline and improve cross-functional communication. For example, at a scheduled time before a company’s monthly S&OP meeting, an AI chatbot could be set up to contact a sales leader to follow up on action items, note deltas between sales actuals and forecasts, and ask for key inputs. The chatbot could follow a similar process with demand planners, supply planners, finance leaders, and other S&OP stakeholders. During the monthly S&OP meeting, the chatbot transcribes the discussion, summarizes takeaways, assigns action items, and automates follow-up. 

Notably, none of these innovations are a full-scale automation or replacement for human involvement—at least not yet. But these are potentially impactful ways AI can increase speed, accuracy, and predictability in an S&OP process that must navigate increasingly volatile supply chains. 

Beyond S&OP, short-term execution—sometimes called “sales and operations execution (S&OE)”—can potentially offer use cases for full AI automation. A large food and beverage company with big historical sales data sets could, in theory, use AI to fully automate same-day deliveries in experimental markets, and then extend the pilot to mature markets once the concept is proven. 

Adds Deloitte: “For longer-range planning at the aggregate category level, we suggest organizations apply more human intelligence.”