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.

AI-era supply chain management needs faster decisions

Improve your speed through better management.

[Photo: Getty Images]

Profile photo of James Barlow

President · ORI

During the COVID-19 pandemic, one of the largest wine and spirits firms in the U.S. quickly realized its typical monthly sales and operations planning (S&OP) cycle was too slow and rigid to respond to daily, even hourly, supply chain impacts.

Using technology to analyze supply and demand signals in near real time, the firm was able to pivot its business model toward home deliveries and avoid existential economic harm.

Six years later, most companies are facing unprecedented challenges in managing highly volatile supply chains, driven by geopolitical turmoil, extreme weather, inflation, and trade wars. As global supply chains grow more volatile and dynamic, product manufacturers and distributors need to adapt their operations and tech stack for a more tumultuous baseline.

Naturally, AI presents generational opportunities if you can sift through the hype to find the right tools, use cases, and workflows. In a timeline where the pressure to use AI for AI’s sake is real, a little precision and restraint goes a long way. For supply chain management, there’s considerable risk that AI is solving the wrong bottleneck.

Many companies now have numerous dashboards, alerts, AI agents, and AI-suggested actions—all of which drive supply chain insights. Yet, unlike the wine company that used data and technology to enable dynamic daily planning, most supply chain teams still make decisions on traditional monthly cadences, and insight overload could delay decision-making rather than accelerate it. The real value is in speeding up the time to a decision.

A means to an end

Sometimes a concept can appear obvious in the abstract but be elusive in practice. The difference between insights and decisions is one such topic.

Insights are signals, patterns, and data—all things that indicate that a risk is present, a forecast is changing, or supply and demand don’t agree. If you see that sales are trending downward in recent months but the forecast is trending upward, or vice versa, this signals your planners are moving the forecast in the wrong direction given the trend.

We call this a “trend break,” and unless the planners have a specific reason to explain it (such as a new product launch or holiday promotion), there shouldn’t be a break in the trend.

Insights help you focus on decisions that move the needle, such as higher product volumes or low predictability in a particular product line. Decisions are cross-functional commitments to take actions on insights, with tradeoffs evaluated and owners and timelines assigned.

In this framing, the insight must be paired with a decision. Furthermore, the value of any insight lies in how useful it is in making better or faster decisions (ideally both).

The software era brought a proliferation of data access and visual management tools into supply chain management and sales and operations planning (S&OP), or cross-functional planning meetings where sales, supply, and finance reconcile forecasts.

For example, the cited wine company used software built on well-structured data to drive continuous alignment and surface high-value actionable insights as COVID unfolded hour by hour. That kind of business value is a far cry from some S&OP contributors who see a beautiful dashboard as the end, not the means.  

Why “speed to insight” falls short

Given the complexity of the supply chain, it’s reasonable to assume that an increased volume of AI insights will naturally lead to better, faster decisions. This logic breaks down quickly in real life for several reasons:

  • Most supply chain decisions are cross-functional: A change in the forecast comes from the planning team, and the response is led by the sales, supply chain, finance, or customer success organizations. Paradoxically, more insights can cripple action through “analysis paralysis.”

  • Integrated business planning data is often fragmented and hard to access: A good, efficient decision may require financial assumptions, exceptions, operational constraints, customer-interaction context, and other data that live in various silos.

  • Companies often make decisions outside their insight systems: Insights appear in a software tool, are transferred to slides, and delivered in a monthly S&OP meeting where decisions are made and captured separately.

  • Volatility ups the stakes: Inaction’s penalty grows as the distance between insight and action widens. When the world moves fast and insights arrive in real time, decision speed must follow for any data to matter.

The core capability of decision speed

The best companies use AI to absorb and filter new information, simulate options, quickly align stakeholders, and commit to action while opportunity windows are still open.

The best supply chain teams use shared dashboards, real-time numbers, and workflows to move quickly and efficiently from analysis to action. The shift to home delivery of wines early in the pandemic resulted from the surfacing of real-time, high-quality data, enabling confident decisions to be made quickly and efficiently.

Three core areas allow AI to supercharge how fast teams decide and act in supply chain management.

1 Move from reactive planning to proactive planning:
Use AI to run effective simulations, generate reliable predictions, and improve forecasting.

2. Move from monthly to continuous alignment:
The S&OP process doesn’t have to be limited to monthly cadences. Dynamic alignment and fluid decision-making are critical as supply chain volatility increases.

3. Move from reporting to decision support:
At its worst, the S&OP process devolves into status updates and reporting with precious few solutions and decision-making. Workflow can lead to faster decisions if AI helps connect disparate data sources and sort and filter data for relevance.

Every supply chain leader should ask these questions about AI right now. Can we:

  • Make a decision based on an insight in the same environment where the insight appears?
  • Simulate trade-offs in real time with the right people and data in the room? (Physically or virtually.)
  • Move from new information to aligned action in hours or days, not weeks or months?

In the AI era, insight becomes cheaper, faster, and increasingly commoditized. Decision-making is still hard, and getting better and faster at decisions is your differentiator.

James Barlow is the president of ORI.