Supply Chain Shifts Collide: Nearshoring, AI, and Automation Converge

AI Is Making Decisions in Your Supply Chain Right Now — Do You Know Which Ones? Most executives in logistics, procurement, and operations aren't asking "should we use AI?" anymore. That ship sailed. The real question — the one keeping supply chain leaders up at night — is: where do you trust it, and where do you don't? Because getting that wrong isn't a tech problem. It's a margin problem. A resilience problem. A career problem. "AI doesn't fail loudly....it fails quietly" Autonomous decisions in planning, inventory, and procurement can compound errors before anyone notices. The executives who are ahead aren't using less AI — they're using it with governance frameworks that give them visibility and control at the right moments. "Resilience isn't expensive.....unplanned disruption is" Anti-fragile supply chains don't just survive volatility. They learn from it. Diversification, scenario modeling, and modular network design are the tools leaders at $500M+ companies are using to turn disruption into structural advantage — without killing margins. "ESG reporting is costing you money. ESG strategy is making it." There's a gap between companies drowning in compliance and those using traceability and sustainable procurement as a competitive lever. End-to-end visibility isn't just a reporting tool — it's a brand advantage and a cost reduction engine. "Your digital supply chain is only as strong as the team running it" AI adoption stalls when the talent model doesn't evolve with it. Reskilling, cross-functional collaboration, and digital-first recruitment aren't HR talking points — they're operational necessities for leaders building supply chains that actually work in 2026. "The best insights don't come from consultants — they come from peers" 90+ senior executives. All targeting the same problems. Companies ranging from $500M to $300B+ in annual revenue. When you put those people in a room — from logistics, operations, transportation, procurement — the conversations don't stay theoretical for long. The supply chain leaders who will define the next five years aren't waiting for perfect conditions. They're building the governance, the networks, and the talent models right now — while others are still debating whether AI is ready. The question isn't whether your supply chain needs to evolve. It's whether you'll be in the room where that conversation happens first. 

77% of supply chain organisations have no formal AI strategy. Not a draft. Not a roadmap. Not even a one-pager. Just a growing list of projects that don't connect to each other or to the rest of the business Gartner surveyed 120 supply chain leaders who had already deployed AI. The majority were operating project-by-project, chasing short-term wins while the technology they were buying accelerated around them. The result? Fragmented systems that can't talk to each other. Spend that converts to cost. And competitors executing against roadmaps, pulling further ahead every single month. Here's what makes that worse: By 2030, half of all supply chain management solutions will deploy autonomous AI agents. The organisations without a strategy today aren't just behind, they're building debt that compounds. A minimum viable AI strategy is six elements. One document. Completable in under two weeks. Most teams don't have that. But here's the truth nobody puts in the infographic: Writing the strategy is the easiest part. The hard part is what comes after the document: Getting leadership aligned on three use cases when everyone has a favourite. Confronting the data quality problems the strategy will surface. Holding the 90-day pilot to its scope when pressure builds to expand it. Naming the owner, and giving them real authority to say no. Organisations pulling ahead with AI didn't just write a better document. They built the discipline, the accountability, and the structure to execute against it and kept executing when it got uncomfortable. That's the gap. Not the page. What comes after it. Does your supply chain have a strategy, or just a list of projects? #SupplyChain #AI #SupplyChainAI #DigitalTransformation 

AI in Procurement isn’t about replacing people — it’s about finally unlocking the function’s full potential. For years, procurement has been expected to drive savings, manage risk, and enable the business… often with limited visibility and manual processes. AI is changing that equation. What I’m seeing (and where the real value is emerging): • Spend intelligence at scale – AI can classify and normalize line-item data in ways traditional tools never could, giving leaders true visibility into where money is going • Market intelligence on demand – real-time insights into pricing trends, supplier benchmarks, and category dynamics that were previously fragmented, outdated, or locked behind static reports • Proactive risk management – from contract deviations to supplier signals, AI can surface issues before they become audit findings or business disruptions • Faster, smarter sourcing – drafting RFx events, analyzing bids, and identifying negotiation levers in a fraction of the time • Cycle time compression – intake to contract is no longer a weeks-long process when workflows are augmented with AI-driven insights • Stronger stakeholder alignment – AI helps translate procurement data into business outcomes leaders actually care about But here’s the reality: AI won’t fix a broken procurement model. If your policies are unclear, your data is fragmented, or your governance is inconsistent, AI will simply accelerate the noise. The organizations that win will be the ones that: 1) Have disciplined processes and clean data foundations 2) Align procurement to business outcomes, not just cost savings 3) Use AI as an augmentation layer — not a shortcut Procurement has always had the potential to be a strategic lever. AI just removes the excuses. Curious how others are applying AI in procurement — where are you seeing real impact vs. hype? #Procurement #StrategicSourcing #SupplyChain #ArtificialIntelligence #AI #DigitalTransformation #SpendAnalytics #CategoryManagement #Sourcing #ProcurementLeaders #CPO #ThirdPartyRisk #VendorManagement #FinServ #EnterpriseTransformation

Most mid-market AI spend is just procurement cosplay. A slick demo. A fear headline. A board member asking, “So… what’s our AI plan?” And suddenly you’re approving budget for vibes. Here’s the question I ask before approving anything: “What’s the first measurable metric this will move—within 90 days?” Not “innovation.” Not “transformation.” Not “keeping up.” A number. A baseline. A target. An owner. If you can’t answer that, you’re not buying AI. You’re buying anxiety relief. The first metric I demand (pick one, no mixing): Cycle time: “We cut customer response time from 18 hours to 6.” Cost per unit: “AP processing cost drops from $X/invoice to $Y.” Revenue velocity: “Quote-to-close time improves by 20%.” Error rate: “Returns/defects down by 15%.” Deflection: “Support tickets reduced by 10% without CSAT dropping.” Cash: “DSO improves by 5 days.” And here’s the trap: If the “metric” is time saved, I want it tied to capacity reallocated or overtime reduced. Otherwise “time saved” is just time… spent elsewhere. CEO rule: If the vendor can’t agree to a metric in writing, with a 90-day proof window? They’re selling story, not outcomes. Either your AI pilot moves a measurable lever fast… or it dies. Publicly. What’s your first measurable metric—before you approve anything AI? #CEO #AI #OperatingModel #Leadership #MidMarket #Execution #Governance #LawrenceLerner OneAccord

AI isn’t scaring markets. Uncertainty about execution is. If you zoom out from stock charts and look at enterprise data, the pattern is consistent across major 2025–26 reports: • Only 12% of 4,454 CEOs say AI has delivered both cost and revenue impact (PwC). • Just 11% of organisations have agentic AI in production (Deloitte). • 56% of US firms say technical debt is blocking new investment (KPMG). • 10–25% EBITDA gains are achievable — but only when governance and workflows are redesigned before tool deployment (Bain). • Firms concentrating on fewer, high-impact initiatives outperform broad AI portfolios (BCG). That’s not a bubble. That’s an execution bottleneck wearing a strategy costume. Markets are pricing future AI productivity at scale. Meanwhile, inside many enterprises: • Annual funding cycles still rule • Data ownership is a contact sport • Governance has more layers than enterprise lasagna • And automation is being politely placed on top of broken workflows Volatility isn’t about whether AI matters. It’s about whether organisations can convert AI enthusiasm into disciplined operating change. The companies that simplify execution, reduce technical debt, and redesign work before scaling tools will compound advantage. Everyone else will continue to produce very sophisticated pilots. The 2026 planning question isn’t: “How much are we investing in AI?” It’s: “Have we re-architected the system that’s supposed to absorb it?”

Stop Buying AI. Start Challenging Your Business Model. After 25 years in the industry—moving from the front lines into leadership—I’ve seen every "digital revolution" promised to us. Most fail. Why? Because we keep trying to pave over cow paths. We take a broken, bloated process and try to "digitize" it. That’s not transformation; that’s just making your mess more expensive. The Mindset Shift: Critical Thinking Over Tech-Buying: The real bottleneck in 2026 isn't computing power. It’s our refusal to think "unusually." We are stuck in a mindset of incremental improvement when we should be looking at radical elimination. If we want to solve truly complicated problems, we have to stop asking how to make a process better and start asking why that process even exists. 1. Finance: The 10% Reality Most Finance departments are drowning in "overwhelming processes"—manual reconciliations, multi-layered approvals, and endless data entry. • The Reality: We can execute finance activities with less than 10% of the current workforce. * The Lever: Agentic AI. Not just a chatbot that summarizes a PDF, but autonomous agents that act, verify, and close books in real-time. This isn’t about "cutting heads"; it’s about converting administrative fat into shareholder value. 2. HR: Trimming the "Fat" to Find the Talent HR has become a maze of operations and "talent management" paperwork. By deploying Agentic AI to handle the operational heavy lifting, we can reduce the workforce requirements in these departments by more than 50%. This allows the remaining human element to focus on what actually matters: Culture and Strategy. 3. Performance-Driven, Not Technology-Driven The biggest mistake leaders make is being Technology-Driven. Buying a new system just because it’s "the latest" is a trap. To drive true efficiency, AI must be Performance-Driven. We must use it to solve the "Business Driver"—reducing costs to the absolute minimum while maintaining the highest levels of availability, reliability, and quality across the board. The goal is to make these operations an invisible heartbeat of the company, not a line-item headache. The Bottom Line Any time a new technology appears, the instinct is to buy it. That is a loser’s game. Real leadership is understanding the business driver first. If you don't have the mindset to challenge the "way we've always done it," no amount of AI will save your margins. Transformation is a "message of hope and realism." The realism is that the old ways are dying. The hope is that by rising to meet these challenges with a fresh spirit and a lean mindset, we build something far more resilient. Let's stop talking about "features" and start talking about "Value." #DigitalTransformation #AgenticAI #Leadership #ShareholderValue #MindsetShift #BusinessEfficiency

Procurement is no longer just about the "lowest price." 🚀 The shift from back-office function to strategic powerhouse is happening right now, powered by AI. If you aren't leveraging data, you’re leaving resilience on the table. Here is how AI is rewriting the sourcing playbook: 🔍 Real-Time Visibility: Moving from reactive fixes to proactive risk mitigation with 360-degree supplier monitoring. 📈 Predictive Sourcing: Using AI models to forecast commodity price shifts before you hit the negotiation table. ⚖️ Smarter Contracts: Automating clause extraction and compliance so you never miss a milestone or a breach. 💡 Innovation Scouting: Identifying "future-ready" suppliers that align with your sustainability and R&D goals. The Bottom Line: AI doesn’t replace the procurement professional; it gives them the "superpowers" to drive real enterprise value. How is your team integrating AI this year? Let's discuss in the comments. 👇 #AIProcurement #SupplyChain #StrategicSourcing #DigitalTransformation #FutureOfWork

Many organisations invest heavily on the tech with little focus on operational aspects, this leads to fragmentation and compromises value #AITransformation

Aleks Sidorecs' analysis of the Anthropic labour market report nails why agentic AI stalls in practice. The gap between what models can do and what organisations let them do is an operating model problem, not a technology problem.                                                                                                                                                                                           I've been rethinking the operating model for data pipelines. The structural parallels are striking.                                                                                                                                                           Aleks introduces the "agent steward" for process ownership. Data pipelines need the same: knowledge about how a source should be interpreted can't live in whoever built the integration. It needs to be a declarative artefact the pipeline executes against.                                                                                                                                                                                 He proposes layered governance: synchronous approval, bounded autonomy, retrospective review. Data pipelines need identical layers. Some extractions run mechanically within declared rules, some hit  ambiguity requiring escalation, some need humans reviewing what the system did. Most pipelines today have one mode: code runs, someone notices when it breaks.                                                                                                        He writes: "An undefined autonomy boundary isn't freedom. It's ambiguity that will resolve itself at the worst possible moment." Pipelines without explicit contracts do exactly this, making implicit decisions about how to interpret headers, map values, handle format variations. Nobody specified the boundary, so the boundary is whatever the code happens to do.                                                                                                        But one line stays with me more than any framework. After two decades of supply chain work across mergers, greenfield builds and turnarounds: "The pattern is always the same: the technology arrives  before the organisation is ready to absorb it."                                                                                                        That pattern holds for data. The models can extract and transform with remarkable capability. The organisations consuming that data have no operating model for it. The good news: recognising this as an operating model problem is itself the unlock. Make the implicit explicit, define the boundaries, learn from what the system does, widen as confidence grows. That discipline works whether you're redesigning how agents make decisions or how pipelines produce the inputs those agents depend on. The organisations solving one are building the muscle to solve the  other. 

𝗥.𝗜.𝗣. 𝗠𝗰𝗞𝗶𝗻𝘀𝗲𝘆. 𝟭𝟬 𝗔𝗜 𝗽𝗿𝗼𝗺𝗽𝘁𝘀 𝘁𝗵𝗮𝘁 𝗿𝗲𝗽𝗹𝗮𝗰𝗲 $𝟱𝟬𝟬/𝗵𝗼𝘂𝗿 𝘀𝘂𝗽𝗽𝗹𝘆 𝗰𝗵𝗮𝗶𝗻 𝗰𝗼𝗻𝘀𝘂𝗹𝘁𝗮𝗻𝘁𝘀. Companies pay consultants thousands for frameworks you can build yourself. Here are 10 prompts I use: → Supplier Risk Assessment → Spend Analysis → RFQ Strategy → Negotiation Playbook → Category Strategy → Demand Forecasting → Contract Risk Review → Inventory Optimisation → ESG Supplier Scorecard → Make vs Buy Analysis Each one gives you a structured output that used to cost a full engagement. The consultants aren't smarter. They just have better frameworks. Now you do too. Follow Asmaa Gad + Supply Chain AI Pro for daily AI tips in supply chain #SupplyChain #Procurement #AI #GenAI #DigitalTransformation