AI, analytics and agility: How modern supply chains win despite increased uncertainty

• Leading firms are turning supply chain resilience into competitive advantage by embracing uncertainty and using advanced analytics to react faster to shifts in supply and demand.
• AI is finally having an impact by boosting decision-making speed and accuracy to gain efficiencies, reduce operating costs and achieve superior service levels.
• Companies that treat supply chain data as a core asset and enabler of competitive advantage can achieve greater agility and reinvention at scale.

In 2025, supply chain organizations face an unprecedented combination of global supply disruptions, technological advancement, structural macroeconomic changes and accelerating technology innovation. In the face of intensifying geopolitical tension and increasingly frequent shifts in trade and regulatory policy, supply chain leaders must navigate a landscape where disruption is the norm. At the same time, AI represents an exciting opportunity to rethink decades-old ways of working and achieve previously unachievable levels of productivity. These trends are leading to fundamental shifts in how global commerce operates.

Insights from the 2025 Gartner Supply Chain Symposium reveal organizations moving beyond viewing resilience as an operational objective to thinking of sophisticated supply chain management as a growth strategy. Industry leaders can gain competitive advantage and increase market share by making better decisions when faced with disruption. This shift in mindset—from viewing uncertainty as a threat to embracing it as an opportunity—marks a critical evolution in supply chain strategy. Companies that effectively navigate uncertainty and outmaneuver competitors with technology, data and cross-functional collaboration view supply chain management as a profitability driver.

One of the central themes from the Gartner symposium was: uncertainty isn’t just the norm—it’s the opportunity. Gartner’s phrase, “the endeavor normal,” reframes disruption as a proving ground for competitive advantage. According to ZS’s Jim Lee, being “reinvention-ready” is more critical than being disruption-proof. Companies that accept volatility as a strategic opportunity—not a temporary inconvenience—are best positioned to lead. In this discussion, Lee joins fellow ZS supply chain experts John DeSarbo and Caglar Ozdag. They explore the current state of supply chain management, emerging trends in supply chain technology adoption, and strategies for building sustainable competitive advantage in an era of constant change.

John DeSarbo (JD): I think there are two ways supply chain organizations are building resilience. One is through strategic moves that build agility such as onshoring or near-shoring manufacturing operations or distribution network redesign. These changes take time but add flexibility when faced with disruptions in supply or sudden shifts in demand. The other approach is more tactical—using technology to reduce decision latency and enable changes that can be implemented faster, such as changes to pricing or inventory levels. These two approaches are not new, but opportunities to execute and implement these ideas have never been more achievable.

Caglar Ozdag (CO): I try to remind our clients not to panic. Disruptions aren’t new. What’s different is today’s technology where the ultimate goal is not to capture the cause of the disruption but how you respond by eliminating surprises. Creating visibility, having a plan B, C and D ready and designing for agility are key. We prefer data-driven approaches—use your data to assess capabilities and failure thresholds continuously. End-to-end simulations let you model disruptions in advance, making you realize when and where you would fail. Once you have that, you can enhance signals with internal and external sensing to get early warnings and execute your predefined response.

Jim Lee (JL): In the past, disruptions were predominantly event-driven—geopolitical wars and public health crises. Today, businesses are dealing with macroeconomic trends such as deglobalization, structural labor changes and technology disruption with the massive impacts of AI. These trends fundamentally alter how businesses operate. And now that disruptions are more commonplace, it’s less about being disruption-ready and more about being reinvention-ready.

JD: The source of disruption is different now because much of it is being driven by changes in governmental policy. Unfortunately, over the past decade plus we have learned a lot about how to deal with disruption caused by public health crises, climate change, war, etc. And we continue to see the frequency of disruptions from these challenges accelerating. All kinds of studies talk about how often a company should expect to encounter a major disruption every two to four years at minimum, and that’s speeding up. More recently, however, the volatility has been rooted in changes to how we do business across borders—for example, structural changes to trade policy, fiscal policy and regulations. The tricky thing is that with all the talk about changes to global commerce, governments haven’t yet settled into the actual changes proposed. There is plenty of “back and forth” debates, but it’s hard to predict exactly where we’re going to be in a year or two and how to plan for the new normal.

CO: There are fewer supply chain resources now, and companies in industries such as life sciences have to compete with traditionally fast-moving industries for them. This shift has exposed long-standing vulnerabilities, such as regulatory delays and underinvestment in advanced capabilities. The chip shortage during the pandemic highlighted how other sectors could adapt more quickly than medical device manufacturers. And with global drug shortages at record levels, modern, resilient supply chain infrastructure will be critical for life sciences.
 
To be disruption-ready, supply chains need to start with connected data. The more you integrate external and internal data, the better your visibility and insights. That foundation supports everything else from advanced tools to faster responses to stronger resilience. Without solid, connected data, it’s hard to build anything that truly helps you effectively manage disruptions.

JD: If you define a disruption as a sudden imbalance between supply and demand, to be disruption-ready you need to be able to coordinate corrective actions across the business. Sales and marketing actions need to be in sync with supply chain and manufacturing actions. When faced with insufficient supply to meet demand, for whatever the reason, it is critical to coordinate pricing, product allocation, inventory management, production and sourcing decisions to ensure customer experience is not affected. This requires intense cross-functional collaboration. A siloed, fragmented approach leads to delays and suboptimal decisions. The key to moving fast as a business when facing a disruption is processes and governance that cut across functional teams and technology that makes this possible.

JL: One area where leaders should focus more is capital. Whether it’s human capital, fixed capital or technology capital, getting more output with what you have currently, or even with less, will be a key theme in the coming years.

JD: Most focus on supply risk. But unexpected demand drivers—shifts in customer behavior and preference due to factors such as changes to regulation—can be just as disruptive. Trade policy and tariffs are getting a lot of the attention now, but changes to environmental and healthcare policy will also affect demand patterns across industries. And it will be interesting to see how foreign exchange rates and other types of tax policy changes will affect demand.

CO: A major strategic gap is the global network’s complexity and cost. It’s hard to trace the full value chain—where things are made and who supplies them. Any geopolitical shift or shortage in critical materials exposes weaknesses few see coming, so visibility efforts should focus on response success rate, not only the causes of disruption. 

JD: One example that comes to mind: We’re seeing a lot of supply chain teams successfully use external, publicly available data and generative AI to improve disruption sensing. And we are seeing more advanced prediction and simulation capabilities enabled by AI that improve scenario planning. So, we are seeing AI help identify disruptions faster and improve risk mitigation through more sophisticated planning.

CO: AI will help. I expected more buzz around generative AI at the symposium, but it honestly wasn’t there. ZS has been focused on AI and advanced analytics for years, and we know how hard-to-achieve but promising the opportunities are. Conceptually, generative AI can do a lot—but making it work takes effort and deep domain understanding. In such a situation, data quality, data connectivity, process controls and AI model development are a mandatory combination, not a pick and choose menu.

JD: There’s growing frustration. Many business leaders aren’t seeing ROI. Same old story: Wrong problem focus, adoption challenges and messy data. AI isn’t a silver bullet. A recent article in “The Economist” dubbed this the “trough of disillusionment,” reflecting how inflated expectations around AI are colliding with the gritty realities of deployment and data debt.

CO: AI works across the board, but inventory is the fastest value driver. Forecasting is another quick win. Value proves out quickly, but the impact on profitability depends on consumption cycles.

JL: Short-term value sits more within supply chain execution: executing production, pick-pack-ship, loading trucks and the like. These actions occur daily, so AI-enabled improvements show up quickly. These areas of operation—fast-moving, repeatable and quantifiable—are where AI’s value becomes undeniably clear within weeks, not months.

JD: AI and analytics can produce better inventory, transportation and labor decisions that in turn reduce costs, prevent stockouts and optimize human resources. This is happening across industries. 

CO: Supply chain data is fragmented. Just creating product data involves multiple sources: dimensions, origin, tax structure and borders. Add Internet of Things and sensor data, and it gets even messier. And then consider the 16 companies between you and your supplier’s supplier. Investing in connectivity helps with creating the visibility that an analytics layer can sit on top of.

JL: Many companies are considering or are in the middle of a multiyear data migration journey. That slows things down, but it shouldn’t. AI doesn’t need to wait, as you can cleanse and migrate data while you transform with AI. Rather than waiting, companies are taking agile, parallel approaches—cleaning and migrating data while deploying AI capabilities in targeted areas. It’s not a waterfall. It’s a flywheel.

JD: Super important. Many organizations have worked to align commercial and operational decisions, but there’s more to do. Even more important is collaborating across the value chain. COVID showed what’s possible—think about how suppliers, manufacturers and distributors accelerated vaccine delivery. That same collaboration will be critical for future structural shifts. 

JD: After COVID, the supply chain earned a seat at the table—because of the crisis. This time, the opportunity is growth. If I manage uncertainty better, I gain share. Supply chain becomes a profit center, not a cost center.

JL: Labor continues to be a pervasive topic. Moving forward, responsibilities will move between employees, algorithms and equipment. In industries like pharma or consumer packaged goods, where supply chain operations often represent up to 50% of a company’s total headcount, automation is no longer optional. It’s about allowing machines to handle repeatable tasks, algorithms to apply data-driven intelligence and employees to manage resources with the right context. It’s less about replacement and more about reallocation.