Background

“Complex selling” is a sales approach used when deals are high-value, involve multiple decision-makers, take longer to close, and require tailored solutions. The focus is on navigating multiple stakeholders to say “YES” in longer sales cycles. There is usually a political element, and the technique requires both Art and Science. In the deployment of complex selling techniques for supply chain planning, we often labeled the Information Technology (IT) group as “Dr. NO.” I taught the teams how to neutralize Information Technology Groups (IT) in the complex sale.

Today, if I were teaching complex selling to Native AI and the new generation of AI platform vendors, I would label the supply chain team and the supply chain planners as “Dr. No”. Sadly, I see that many of these teams, steeped in traditional thinking, have become the blockers, not enablers, to adopting new ways of work and embracing new forms of tech.

The Winding Path

The supply chain is a complex, non-linear, distributed system that must be adaptive to meet business needs in an uncertain world. Unfortunately, today, most deployed technologies amplify and distort the signal to improve reliability and drive growth in a world that is becoming more uncertain. Companies talk about the symptoms with a lot of handwaving and acronyms, but don’t know how to roll up their sleeves and embrace uncertainty.

Closing the performance gap is seldom a single factor. The answer is not a single technology or a simple change of approach. While business users are enamored with AI hype and experimenting with agents and agentics, embracing the full value of artificial intelligence requires an AI-enabled architecture, from data architecture definition to solution delivery. Most clients I work with do not know how to get started and fall prey to the AI-stupid pitches from technologists.

Most supply chains have multiple flows—an efficient flow that can be managed at the lowest cost, an agile flow that requires a focus on reliability of cost, quality, and customer service despite variability and uncertainty, and a responsive flow that requires a focus on short cycles. Prior solutions were unable to recognize and manage the flows and automate the rules for customer and product segmentation. As a result, there were many workarounds, black holes, and spreadsheet dependencies. The use of these new approaches recognizes flow and automates the rules. They close the gaps in today’s solutions that require custom code, workarounds, and spreadsheets.

While many social influencers push different narratives emphasizing  “autonomous,” “self-healing,” “real-time planning, “” continuous planning,” or “self-driving” planning, this is not me. I try to sidestep this hubris and hype. My goal is to help companies understand real and tangible use cases to redefine/improve work.

The World is Grayer and Less Certain. Companies Have Not Adapted.

Over time, in my role as an analyst, organizations have become more fragmented, with few companies having purchasing, distribution, and manufacturing reporting to a common leader. We find that reporting relationships matter in the analysis of the Supply Chains to Admire.

We find each leader purchasing systems for their own functional silos, creating a barrier to interoperability. The opportunity is to redefine the architecture to drive interoperability and drive holistic workflows, better aligning the organization with value-based delivery. (The measurement of functional objectives tied to bonus incentives creates waste. Companies need to align to drive reliability.)

Figure 1. Supply Chain Organizational Definition

Today, there are more unknowns than knowns. The traditional approaches to supply chain decision-making focused on known inputs, using known models to drive known outcomes. In the world of unknowns, generative AI, agent-based workflows, What-if Simulation, and What-if Optimization are growing in importance. The benefit is quicker, role-based insight, along with workflow collaboration, to answer questions, increase awareness, and drive action.

Traditional APS solutions focused on known inputs, known models, and known outputs. I laughed this week with the publication of the Gartner Magic Quadrant. The Reason? Fifty percent of the solutions listed as top performers with both vision and execution do not scale for the global multi-national. Most focus on optimization in an architecture that is only a good fit for smaller, regional teams. By and large, we are not asking the right questions.

A New Dr. No Is In Town

The buyer today for supply chain planning is more conservative. The leaders — Chief Supply Chain Officers — are hardened and conservative, with many becoming “Dr. Nos” during sales cycles while pushing traditional definitions of technology. They have not invested in reskilling to learn new concepts and sort reality from hype.

The reason? The deployment of new approaches requires the learning of a new language, rethinking first principles, and being open to new outcomes, all of which fly in the face of tradition. In Figure 2, when potential buyers of Supply Chain Planning were asked the question, “In your opinion, what is your company’s preference in the purchase of new technologies? Would you classify your company as an innovator, early adopter, mainstream adopter, late-stage adopter, or laggard?  The market shows marked change. The innovators and early adopters accounted for 42% of the market in 2016 and 24% in late 2025. When tested for demographic consistency, this result is significant at an 80% confidence level.

Don’t you find it interesting that innovation at scale is the new reality, but organizations are more cautious?

Figure 2. Shift in Innovators

So, how is it that answers at scale and relevancy are now available, and companies have become less innovative in using new technologies? It often comes down to the fact that companies assume historic practices are best practices, and they have never questioned the current state defined by technologies from four decades ago.

If you are a business user, look around your organization for the Dr. Nos. You will know them when they embrace historic practices as best practices, or when they spout hype-based word salad from their tongues. They are usually the members of the team who get excited about the use of agents in spreadsheets or are quick to buy based on the Gartner Magic Quadrant.

The Path Forward

As a business leader, in your talks with your analytics teams, the path forward is to redefine the architecture with the goal in mind from the data up, with a focus on value.

Measure and Understand Variability. To understand the world of possibilities, make a list of supply chain issues and data available–all forms structured, streaming, audio, text, images–and make a visit to your data science teams and explore the building of a semantic layer with a graph (for relationship and semantic reconciliation), along with an ontology to drive flexibility in decision support architectures. Layer this underneath your current architectures to gain insights into planning master data — inputs into planning systems that are variables but treated as constants. These include lead times, conversion rates, yields, price, run rates, etc. Use machine learning and pattern recognition techniques to understand the impact of these inputs on plan outputs.

Define and Measure a Good Plan. As you analyze your plan, focus on the balanced scorecard metrics of growth, operating margin (not cost), inventory turns, and Return on Capital Employed (ROCE). Then measure, with the help of your data science teams, the reliability of the plan by analyzing:

• Forecast Value Added (FVA). An analysis of the value of demand-stream forecasting based on forecastability and flow characteristics.
• Inventory Health by Form and Function of Inventory with an Analysis of Inventory Value Added (IVA).
• Raw Material Value Added. How much improvement did the plan make in the purchase and storage of raw materials? (RVA)
• Customer service: On-time and In-full measurements with reason codes.
• Schedule adherence. The adherence of the manufacturing teams to production plans.
• Bullwhip Effect.
• The Effectiveness of Demand Shaping Programs: Shaping versus Shifting.
• Asset Utilization

In Figure 3, I share a high-level architecture perspective as a starting point for the data science conversation.

Figure 3. High-level Architecture

Summary

My advice? Embrace innovation to define new opportunities, develop better process capabilities, and redefine work. Side-step the hype and focus on creating value for your firm. Traditional supply chain planning approaches are not sufficient, but the redesign requires crafting a solution with the goal in mind, from the data up to new solutions to unleash new forms of value.

“The man who moves a mountain begins by carrying away small stones.”

Confucius, born Kong Qiu, the revered Chinese philosopher

This quote isn’t about literal mountains—it’s about tackling challenges. Even the most daunting tasks are achievable when tackled gradually, one small effort at a time, turning patience and persistence into remarkable progress.

In this picture, we depict a forward path surrounded by small stones. Your journey as a supply chain leader is a journey of carrying small stones.

Climbing the Mountain

I am not patient, but I am persistent.

My goal is to help supply chain leaders get unstuck. My observation is that despite technological evolution and the promise of capabilities, we are having the same discussions, not realizing that most of the current dialogue centers on groupthink — focused largely on symptoms rather than fixes. Here, in the words of Confucius, I want us to focus on identifying and carrying small stones.

Let’s start by focusing on the mountain the leaders are trying to climb. Simply put, our systems do not know how to talk to each other. It is hard to put data to work without semantic reconciliation, a common data model, and an ontological framework.

Here are two examples. (I can list many. The current white paper that I am writing is capped at twenty-five, but I will not bore you here.)

DRP and TMS Lack a Common Data Model. Companies want to be customer-centric and share order-reliability data, but the lack of a common data model between Distribution Requirements Planning (DRP) and Transportation Management Solutions (TMS) is a barrier. In DRP, lead time is usually set to a constant value. (Time to travel from point A to point B on a lane or a route.)

In reality, lead time is a constantly changing variable in supply. Actual lead times should feed into Available-to-Promise, inventory safety stock calculations, and deployment logic, but it doesn’t. Transportation data stays in logistics, and shipment data stays in DRP. Moving data from one model to another requires a common data model and a planning master data layer. The two solutions are not aligned at a goal or a first-principle level.

Table 1. Shifts in First-Principle Thinking for Supply Chain Planning

As the global multinational evolved, the first-generation planning solutions focused on demand variability, assuming that supply variability would be low. It is not. The new requirement is to embrace the fact that both demand and supply are variable, and to use machine learning on transportation data to help map supply uncertainty.

–Mapping Ship to and Ship From Information. In mapping demand data, using consumption or point-of-sale data, a successful solution needs to bridge the ship-to information (customer distribution network logic) to the ship-from information (supplier distribution network). The answer lies in ontological modeling to map customer and supplier logic, drive semantic reconciliation of the item number and the definition of the saleable unit (selling unit of measure), understand dependent demand (items within a shipper or a package), and the shipping instructions. CRM data cannot be used in Supply Chain planning without translation.

While many use the terms agents, agentic, and agentics interchangeably. They are not the same. Agents are the actors; agentic acts are autonomous decision-makers; and agentics is the construction of systems with agents governed by system design. A focus on agents without resolving semantic reconciliation is a fool’s play.

Take a look at the architecture in Figure 1. This type of architecture is required for AI enablement. An investment in agents without semantic reconciliation has limited value.

Figure 1: Architecture for AI Enablement

Moving Mountains

So, if you are a business leader and feeling stuck (as most are in my discussions), side-step the discussion of agents and ask, “Is this a semantic reconciliation opportunity?” Followed by: “How are we aligning architecture requirements with shifts in first principles?”

Lead by asking the organization to build an architecture that enables artificial intelligence. Current architectures are insufficient. Solutions with our favorite acronyms — CRM, APS, ERP —evolved from schema-on-write architectures with a focus on transactional data.

As business requirements have changed, the answer is not integration, but interoperability. AI enablement requires investment in schema-on-read architectures that support the use of unstructured data and new forms of analytics.

As most of you know, I am a supply chain gal. I cut my teeth believing that the answer to supply chain problems lay in better math. At the beginning of my career, this was the promise of Advanced Planning (APS). Life has taught me to think more broadly. Supply chain professionals are laser-focused on better engines, but architectural interoperability remains a barrier for the industry.

So, how to carry your small stones? Educate your teams. Work with data scientists to understand new capabilities. Ask your lawyers to focus on the interoperability of data models in licensing. Build architectures to enable AI capabilities. Don’t fall prey to vendor hype of agents on top of outdated architectures.

Here Me Speak

If you want to know more and gain a greater understanding, I will be speaking at:

• P44 Executive Workshop on April 9th.
• ASCM Top Management Night in Rochester, NY on May 29th

Definitions

Agentics. Multiple agents working together through governance and workflow to drive autonomous process automation.

Synthetic Data: Synthetic data attempts to preserve the patterns and relationships of real data without exposing the original data.

Retrieval-Augmented Generation (RAG). A technique used in AI systems to make language models more accurate and up-to-date by retrieving external information before generating an answer.

Over dinner, the dialogue was heated and intense. We discussed how the first principles of supply chain planning are changing, and that forward progress was not about AI for AI, but about driving improvement. We both agreed that the need state has changed at a first-principles level with the evolution of the global supply chain in a chaotic world. We discussed how AI techniques can help companies achieve the goals of these shifts by applying first principles to complex, non-linear, distributed systems that need to be adaptive.

Before we realized it, we had closed the restaurant. The waiter quietly and politely escorted us out of the restaurant.

This model is so central to today’s news. Uncertainty reigns.

First principle thinking breaks down a complex problem into its most basic elements. This type of thinking allows teams to identify what is absolutely true and discard assumptions, leading to the development of innovative solutions. The discussion with Joe centered on the details of Table 1.

Note: Bi-directional orchestration enables the matching of demand shaping levers–marketing, advertising, social/digital efforts, price, channel strategies, promotions, and new product launch– with supply strategies of alternate sourcing, choices in bill of materials, alternate manufacturing strategies, postponement, shifts in logistics modes, and late stage differentiation to orchestrate demand across make, source, and deliver. Effective engines sense and shape supply chain flows at the speed of business.

The Prologue

At the end of the dinner, I was exhausted. I had a two-hour drive home. The fall evening was dark and damp, and the headlights were wrapped in a dense fog. Yawning, I slowly navigated the winding roads.

Out of nowhere, during the ride, a large deer jumped across my hood onto my windshield. BOOM! in an instant. The shock deployed the airbags and destroyed the car’s frame. I could barely drive the car home. At 1:00 AM, I crawled out of the back of the SUV at my destination.

The vehicle was totaled. I received a check from the automotive insurance company for 40% of what I paid for the car three years ago, and was forced to buy a new vehicle. Thirty thousand dollars later, I had new wheels.

I liked my old car. It was comfortable. I knew how to use the dash. We had a relationship of sorts. I am slow to bond with my new vehicle. We don’t have a positive relationship. It isn’t that I don’t like the car; it is just too complicated to use.

The new car’s software automation has a lot of bells and whistles I don’t want or need, which delays my user satisfaction. My learning curve is slow, while the capabilities are vast. My new car is safe and smooth to drive, but I hate the experience.

As I looked at cars, I was bewildered by the array of electronics (features that greatly outstripped what I need) and surprised by how this barrier keeps me from loving my new car. I miss the days of knobs and buttons.

The new car also spews false alarms. When this happens, the car, connected to the dealer network, sends mandatory alerts for forced maintenance. However, I cannot see the issue (the codes aren’t visible to the driver, but a generic alert is sent to my phone as a text), nor can the shop tell me what is wrong until the car’s computer is analyzed. I try to ignore the alarms, but the shop is diligent on follow-up, mandating that I return to maintain the warranty. As a result, I have spent three unnecessary afternoons in the repair shop.

Which brings me back to the world of supply chain planning. As I listen to briefing after briefing in which technology providers are excited about new AI-enabled features, I hear many pitches of tech for tech. Amidst many, I want to raise my hand and ask, “Where are we at in the redefinition of the user experience based on design thinking? And what are the first principles of design? And why are we talking about making traditional processes faster rather than driving value? And, how do you define value?”

In essence, we are not questioning the first principles of planning and driving improvement for outcomes or designing to improve the user experience. Which brings me to emphasize the meaning of design thinking:

The Epilogue

So, if you are a leader of business teams implementing supply chain planning, here is some advice. Focus on design thinking and first principles. Push this discussion hard to the point of discomfort for the team, because many have never thought about planning at a first-principles level. Do this before you EVER engage with a technology provider or issue an RFP.

In the purchase and implementation of software, reverse the focus. Today, 70% of a team’s focus is on software selection and implementation, and 30% on adoption. Shift the mindset from software selection to driving value through evolution. Sidestep the bells and whistles, and stop the love affair with shiny objects. Don’t embrace AI for AI.

To drive value, take the following steps:

Making a Decision:

• Governance. Get clear on how to make a decision in the selection process. The selection of software is less important than driving value through deployment. RFPs and demos are less important than testing the available options with your data and your scenarios.
• Focus on First Principle Thinking. Before talking to a technology company, use first-principle thinking to define process flows. Consider supply chain planning as a dot within a set of dots in your enterprise software portfolio. Focus on interoperability and process flows within and through the dots: integration is not sufficient. Train your teams on the differences between integration and interoperability, linear and bi-directional flows, and synchronization and harmonization of data.
• Side Step Shiny Objects. Don’t fall for Technologist mumbo jumbo. For example, there are no end-to-end solutions in the market. However, in the process, don’t undervalue the importance of a unified data model for demand, inventory, deployment, and manufacturing planning. Map these flows to synchronize the underserved areas of procurement and logistics.
• Business Led. The most successful planning implementations are business-led. Don’t let the deployment be led by IT.
• Carefully Select a Boutique Implementor. The greatest success happens when the technology provider partners with an experienced boutique implementor. Sidestep partnerships with larger firms without the software expertise.
• Define Goals. Before you implement and align the teams to answer the questions:
  • What does success look like?
  • What is a good plan?
  • What are our constraints?
  • How frequently should we plan?
  • Governance: Who should plan and who should make final decisions?
  • What should be measured?
  • How do we define roles and work?

Through the Evolution:

• Use Design Thinking to Guide the Implementation. Use design thinking to align functionality with process design and capabilities. Avoid bells and whistles that you do not need. Delay upgrades: let others work out the bugs of releases. Upgrade only when mission-critical.
• Thread Value Through Your Lifecycle. Ask software providers how they deliver value through the lifecycle. This includes conferences, training, benchmarking, and insights on how to design work. Post-implementation, consider hiring a coach for the team (someone familiar with the technology involved) to audit software usage at 3-, 6-, 9-, 12-, 18-, and 24-month intervals. (The timings are not cast in stone, but a recommendation.) For example, this is Reveal’s business model with SAP deployments.
• Train, Train, Train. Focus on the user experience and reward innovation in work process development. Measure value through a balanced scorecard and reward innovators. Build a large language model (LLM) to help planners understand their own data–leadtimes, conversion rates, schedule adherence, quality of conformance, Forecast Value Added (FVA), customer service reason codes, etc. Enable discovery.
• Build a Successful Center of Excellence. Supply chain centers of excellence come in many shapes, forms, and flavors. The most successful focus on the delivery of corporate value. Success varies by cultural differences. Center the group on first principles and design thinking.

The End

At the end of the day, a solution is not successful if it is not used and does not deliver value. Don’t fall prey to the current AI Stupid mindset of layering agents on traditional supply chain architectures. The reason? The first principles that shaped the first generation of planning for small and regional supply chain teams do not serve global multinationals well. The solution is not designed for the current level of uncertainty, nor to drive a great user experience. It is for this reason that 94% of planners use spreadsheets as their primary planning tool, and over 90% of companies are dissatisfied with their supply chain planning deployments. (Source: recent research by Supply Chain Insights.)

The venture capital firms and Private Equity partners are pushing their companies hard to AI Stupid layering agents on top of traditional architectures without rethinking first principles or the user experience. As a business leader, be emphatic and educate the investors, but don’t fall prey to the market hype.

Let me start this blog with a heartfelt thanks to all who have taken the time to complete my surveys, participate in my open training, attend my events, and join my network calls. More than 150 readers sent personal notes via LinkedIn on this anniversary.

I never wanted to start my own company. Being a small business owner is a hassle. One that I will willingly relinquish.

History

Over the last 14 years, I have written 605 blog posts, 13 books, hosted five share groups, and penned more than 100 reports. The posts are based on interviews, qualitative and quantitative benchmarking, client work, case studies, and correlations between a company’s choices and its public corporate reporting. I pride myself on speaking truth to power and advocating for the business innovator who is trying to drive a difference in outcomes.

But, I Failed

Recently, Jeff Metersky, now at Gains Systems, wrote:

“Some industry analysts, pundits, and gurus have been making largely the same observations for years. Similar diagnoses. Similar recommendations. New terminology, some times.
And yet, many of our underlying challenges remain. That raises a fair question. If the guidance is sound, why does so little change at scale? And if there is success from their recommendations and guidance where is it communicated?

I do not begrudge anyone for making a living, and some advice is genuinely useful. Yet, too often the true understanding of available solutions is stale, with limited effort to validate current capabilities or outcomes beyond what vendors claim. So when the same conclusions surface year after year, it is reasonable to ask whether we are learning anything new from this group.

I am closer to nearing the end of a 40-year career in this space. I have had the privilege of helping move the needle inside companies, and for my customers and clients, not just commenting on them.
If everything is truly as broken as it is often described, perhaps more voices should move beyond observation and into execution. Build something. Put capital at risk. Test the ideas in the real world.
Because progress is not proven by commentary. It is proven by outcomes. I hope this is read in the spirit it was intended. To find meaningful ways to make tangible actual progress. Please share your thoughts. Open constructive discussion is good.”

I read the post and winced. Yes, I took it personally. I tried, and yes, Jeff is right. I failed. I, and every other influencer, failed. However, I would add, so have technologists and consultants.

My Backstory

I left the world of software twenty years ago, and I don’t personally write code, but I have collaborated with Elemica/BASF and Evonik to test blockchain, and Kinaxis/OMP, and o9 to test outside-in planning processes. (The most significant adoption of this work is the current definition of o9 Solutions demand planning.) I have written about the testing openly and shared the results through this blog and reports.

I am dissapointed that more of the pilot work stalled and was not adopted.

I wrote the Supply Chains to Admire and collaborated with Georgia Tech to build regression models on value. I am dissapointed that this work never found its way into academic journals.

I shared the learning openly through monthly networking calls, sharing of the testing data through open content research, and focused presentations at events, but I agree, I did not make a difference.

So, let’s answer Jeff’s question of how do we make actual progress? I have thought long and hard on this topic. I accept the challenge. I think that it centers on education, value-based content creation, and first-principle thinking. Today, I am going to focus on education and first-principle thinking.

Educate the Buyer to ask Tough Questions

Currently, the market is wrangling through the hype cycle of Artifical Intelligence (AI) everywhere, but meaningful AI case studies are nowhere. In my effort to educate supply chain leaders, I volunteer to speak.

At the end of May, I am speaking to the ASCM Rochester Chapter event on the future of AI and the redefinition of work. My goal is to help.

In the prep, the head of the chapter shared that prior to my speech, the chapter is hosting a speaker on AI. So, I asked a simple question, “How does the prior speaker define AI?” He could not answer the question. So, I probed, “Will he share case studies of the use of Large Language Models (LLMs), machine learning with deep or reinforcement learning, or agent-based workflow or Agentic governance?”

He stammered, “I don’t know.” An uncomfortable silence filled the space.

I often find this to be the case. We are speaking about AI without clear definitions, applicability to use cases, and the right fit of technology. But, more importantly, without a focus on first-principle thinking. Technology for technology sake is a losing proposition. The autonomous supply chain is a fairy tale.

This is all happening as AI applications like ChatGPT and Microsoft CoPilot are radically changing our personal lives. Grammarly has improved my writing and the AI applications in Shutterstock yield impressive images.

However, I firmly believe that we are stalled when it comes to improving supply chain processes.

Just as digital technologies radically improved our personal lives, digital transformation drove little improvement in the core processes of supply chain planning. Here are some questions that I would ask the technologists that show up with shiny object syndrome of AI everywhere, but nowhere. Fight back when the promise is localized efficiency improvements on outdated processes.

So, what to do?

Ask technologists questions on AI to have more honest discussions:

• Why do you think having the dots makes you capable of connecting the dots?
• How do you define AI, and for you, what defines supply chain excellence?
• What supply chain first principles does your software support?
• If they cannot answer, reframe the question as, What makes you believe you can deliver truth when you cannot even name the first principles of truth?
• Why do you think the Internet is the source of trustworthy data? How do you trust, but verify?
• Why do you believe the regurgitation of history breaks the norms to define unprecedented foresight?
• Why do you call your AI artificial when it purports to tell the truth?
• How does your AI solution provide emotional intelligence, and sentient comprehension?

Next week, mis-guided technologists will share their visions on continuous and real-time planning and autonomous work systems at Manifest, but the audience will ignore the over-arching problem. The issue? Companies are not clear on what defines excellence. How can we automate without clarity?

The journey for supply chain excellence has to start with clear definitions. Excellence is not rooted in the efficient delivery of functional metrics. A focus on functional metrics throws the supply chain out of balance and increases waste.

Which leads me to a question Jeff, “Is our current failure a platform for future success?”

Need for First-Principle Thinking

First-principle thinking reverse-engineers complicated problems and unleashes creative possibilities. The approach breaks down complex problems into basic elements and then reassembles the components from the ground up. Having knowledgable team members with deep systems and design thinking, ignites meaningful discussions. People that think they know the answers drag down innovation.

Elon Musk’s First Principle Reasoning Framework

• What is the problem?
• What do we know to be true?
• What are the obstacles?
• What can I do differently?

So, I asked ChatGPT for the first principles of effective supply chain planning. Here I share a bulleted list from the query:

• Demand drives everything → Pull beats push.
• Flow beats efficiency → OEE ≠ throughput.
• Variability is the enemy → Stability before speed.
• Time is the real cost → Lead time hides waste.
• Inventory is a buffer, not a solution → WIP hides problems.
• Constraints govern the system → The bottleneck rules.
• Information moves faster than product → Signals > schedules.
• Trade-offs are unavoidable → Choose your operating point.
• Reliability beats optimization → Predictable wins.
• People and incentives shape outcomes → Metrics drive behavior.

What do you believe are the first principles of supply chain planning? Are you aligned as a group before you evaluate technology?

As you consider the topic, ask yourself: why do we have supply chain planning systems that prioritize push-based flow to maximize Overall Equipment Effectiveness (OEE) and Efficiency, yet provide little insight into managing lead times and cycles? I think the answer lies in the first principle: people and incentives shape outcomes. Companies do not understand the value proposition of waste and cost resulting from misalignment with first principles.

As supply chain planning evolved, flows between functions were not automated, and processes/bonuses were introduced to optimize functional performance, often at the expense of the balance sheet. Technology was not sufficient to model the supply chain as a complex, non-linear system, so we did the best we could. Unfortunately, we have never returned to address the problems of limited, expensive memory and lengthy batch processing in the early definition of supply chain planning technologies.

Ironically, today, the platform definitions are similar, but faster. Few question the basics and try to align with first principles.

Demonstrating these first principles to traditional leaders in finance, information technology, and manufacturing requires modeling that most supply chain leaders lack the capability to build, but it can be built with consultants and newer technologies like Lyric and Optilogic.

So, Why Are We Failing?

Jeff, I think that the answer lies in the fact that technologists are talking about technology for technology’s sake, while business leaders are rooted in traditional process thinking. Both need to change. The momentum and force for traditional processes are just too strong. No independent voice can lift us past this cacophony.

Technologists are biased toward selling software, and their views shape information feeds and events. We need to side-step this tradition.

To evolve, both groups have to step away from hype cycles and shiny-object syndrome to focus on first principles. This evolution takes a village. Hopefully, you will be willing to help.

Note: Some of the questions about what to ask a technologist about AI were shared with permission from Georges van Hoegaerden, georges@ivanhoeinstitute.com blog (georges@ivanhoeinstitute.com).

“One thing I’ve been thinking about as I read your work and the newsletter is that most leaders already agree the system is broken – but they don’t yet have a low-risk way to act differently.

What seems to be missing isn’t insight; it’s a small, repeatable experiment that helps teams see the mismatch you’re describing.

For example:
– What happens if, for one cycle, a team replaces forecast accuracy with a single outcome metric (customer fulfillment inside lead time, revenue at risk, order stability)?
– Or if they label every planning input with “freshness” before trusting the math?

My hunch is that once people experience the gap between optimized plans and executable outcomes – even in a narrow slice – the unlearning starts naturally.

Curious how you think about converting all this research into a few “safe-to-try” experiments that leaders can run without a major transformation mandate.“

Safe-to-Try Experiments

I think that this is a great callout. The market is a bit of a mess with a lot of hype. AI is everywhere, but nowhere, and supply chain planning technology providers are being opportunistic. Most are touting AI STUPID techniques. Never layer Agents on existing platforms. There is no one RIGHT solution. Your goal is to sort through many flawed solutions.

Until the market sorts itself out, here are my recommendations.

Start by defining your supply chain using the River of Demand Methodology. When insights are unclear, color the water brown; when insights from data are clear, color the water blue. For reference, here I share a River of Demand generated through the OMP Project on outside-in planning, in collaboration with Nestle, for a New Product Launch. List your rocks (barriers), data sources, and roles. As you do this, identify what each role would benefit from seeing.

Then focus on which roles and processes to automate using new forms of technology. List these and, using design thinking, map what each role needs for insights.

After mapping these flows and roles, consider these quick-win projects.

1. Build Customer Listening Posts. Spend the time as the market sorts itself out to get good at analyzing customer sentiment. Use unstructured text mining and Large Language Models (LLMs) to gain insight into true customer sentiment. Analyze return data, damage, on-time performance, customer one-off requests/exceptions, inventory levels, and order patterns. (For consumer-facing companies, include rating and review data, shelf imaging to understand planogram compliance, price, and broker data.) Build a cost-to-serve model and use the insights in top-to-top customer meetings and monthly/quarterly customer listening post reviews.
2. Unified Data Model for Planning. The average company with revenue greater than 5B has seven to twelve planning technologies, each running like a spinning plate. They have a different cadence. The models are not interconnected to drive holistic understanding. To build a semantic reconciliation layer, construct a unified data model, and use Celonis process mining to drive bidirectional process flows between dissimilar models. Write the output to a system of record for organizational sharing and alignment.
3. Train Teams and Build Knowledge on AI Concepts. Build a training program to upskill your teams. Don’t talk AI; instead, speak the language of AI with an understanding of use cases, technology enablers, and infrastructure readiness.
4. Digital Assistant (s). Use technologies from Lyric or Optilogic to build digital assistants for key personnel. Each digital assistant is a listening post to gain role-based insights. Allow roles across the demand river to continuously query supply chain master data and gain insights into questions companies do not yet know to ask. A digital assistant is a native AI network design technology embedded in a large language model to make the design more actionable and enable bi-directional query.
5. Employee Training Using Large Language Models. Build a multilingual training model that enables employees to move data from portals into more usable large language models (LLMs). Consider including definitions of organizational-specific language, acronyms, training materials, supplier and manufacturing capabilities, competitor information, market share insights, etc. Constantly feed the supply chain master data into the training module to keep it up to date. I worked with a company to build an LLM for training, saving them 5M in charges to Accenture for Accenture Academy.
6. Supply Chain Planning Master Data. Build a supply chain planning master data layer to analyze market data. Use machine learning to understand the median/mean values and the variability of inputs to supply chain planning, including lead time, conversion rates, commodity prices, production/manufacturing capabilities (speeds and feeds), supplier reliability, tariff factors/rates, etc. Read, analyze, and transform the data to plan batch processes for each run. Feed the plan and the digital insights layer with the insights.
7. Native AI Supply Chain Design Platforms. Consider replacing your legacy design tools, such as AIMMS, Llamasoft, OMP, and Blue Yonder, with faster, easier-to-use solutions from Lyric and Optilogic. Run these design/what-if analyses with each S&OP cycle to design the supply chain, including parameters such as (but not limited to) buffers, push/pull decoupling points, inventory strategies, policy decisions, inventory run-out plans, and potential sourcing strategies.
8. Sales and Operations Planning Execution. Build a platform to host playbooks (likely scenarios from S&OP discussions). In S&OP, instead of building one plan, build multiple scenarios based on team discussions. To drive an S&OP Execution process, implement a process where a small team calls playbook audibles to make weekly decisions based on tactical planning scenarios. (Much like football teams call plays in a huddle.)
9. Implement Demand Planning based on Demand Flow. The most advanced thinking in this area is within o9 Solutions. Watch this YouTube series for insights and then use the graph to model multiple demand flows using market data. Focus not on error reduction. Instead, measure Forecast Value Added by stream.
10. Focus on Reliability. How Good Is Your Plan? Analyze your current planning processes. Shift the focus from functional metrics to building a balanced scorecard and then evaluate the reliability and planning effectiveness of current outcomes. Here is a list to get started:
    • Focus Metrics/Incentives on a Balanced Scorecard. You are what you measure. Shift bonus incentives from functional metrics to align with a balanced scorecard. The metrics that we find drive the greatest value (market cap/employee through work with Georgia Tech) are Annual Growth, inventory turns, operating margin, customer service (on time and in full), and Return on Invested Capital (ROIC). Measure and compare the patterns of each supply chain to industry potential.Inventory Health. Analyze current inventory health and evaluate safety stock levels using insights from the planning master data layer on lead time variability.
    • Bullwhip. Use network design technologies to develop different scenarios and measure the bullwhip effect. Translate the impact of the options on operating margin, customer service, and growth.
    • Production Plan Adherence. Calculate production plan adherence and understand the deviations.
    • Forecast Value- Added (FVA). Shift from measuring demand error to Forecast Value Added (FVA). Drive continuous improvement programs to improve FVA.
    • MRP Value-Added. Apply a similar process for FVA to MRP. Analyze the value of current MRP practices to a naive material plan.
    • Success of First Pass Tender. One issue with traditional approaches to transportation planning is the focus on the lowest cost and the lack of accountability for a feasible plan. Track carriers that are granted lanes based on cost, and do not pick up loads when assigned to them. Track the impact on costs and customer service, and reassign lanes based on a multifactorial analysis of both cost and carrier reliability.

Here is a metrics hierarchy to consider when driving operational reliability.

Unsolicited Advice

Don’t buy large planning platforms right now. There is too much change happening.

Also, don’t buy software from a consultant. I have never seen a consulting company successful in building and maintaining software.

Instead, try these small projects to maximize the value of current technologies. As you conduct your assessment, consider how you apply the right techniques to solve your problems and drive insights. There is no ONE right answer.

Heuristics are fast, but not deep. Optimization yields solutions to problems you can model with known structures, given a well-defined objective function. Optimization is slower, but deeper than heuristics. In contrast, Machine Learning gives you predictions from data you can’t model analytically (without an explicit model). Most real problems need a combination of these techniques. Put heuristics to work for image recognition, data mining, and pattern recognition while deploying AI within deeper mathematical engines in optimization for faster and deeper answers using reinforcement learning, deep learning, and gradient descent. (The list of techniques is long and evolving.)

Don’t confuse what-if analysis with discrete event simulation. Both are becoming easier to use and deploy. Discrete-event simulation is more difficult, but it is useful to test the feasibility of a plan.

Work with your data scientists to identify problems and align the right techniques to drive insights. When you ask your technology providers how they use these techniques to drive better insights, put a data scientist on your project team and ask the technology team to outline their future roadmap that integrates these approaches.

Don’t myopically focus on engines. Instead, broaden the discussion to include disparate data sets, model types, synchronizing data at different velocities, and building the planning master data and semantic reconciliation layers.

And, stop anyone who mentions the Gartner Magic Quadrant or any other fairy dust framework. They just are not relevant right now.

Summary

So, Vijay, how did I do? Anything to add?

Good luck in your journey.

Y2K was ending when I posed in front of a camera for my Gartner ID card. This year marks my twentieth year as an industry analyst. I spent two years as an analyst at Gartner Group, six years at AMR Research, two years at Altimeter Group, and the remainder as the founder of my own independent analyst company, Supply Chain Insights.

What is an industry analyst? I like to tell folks that a consultant knows the answers, while an industry analyst is searching to formulate the right questions to ask to improve business outcomes. As those who read my blog regularly know, I take this job seriously. I write for business leaders seeking first-mover advantage.

Reflection

I left Gartner because I did not believe in the business model. I watched as ERP providers influenced fellow analysts to push ERP II concepts myopically. I protested, but found my differing views were a hard slog up a long hill. Most operated in an efficient transaction paradigm.

Here we are, twenty-five years later, making little progress in supply chain planning, facing the same issues I cited in our Gartner meeting rooms during analyst discussions. I felt then, and I feel now, that analysts’ perspectives were hijacked by ERP vendors’ spending on analysts and by most analysts’ lack of understanding of supply chain planning.

For reference, here is the Wikipedia summary of ERP II.

“ERP II” was coined in 2000 in an article by Gartner Publications entitled ERP Is Dead—Long Live ERP II.^[15][16] It describes web–based software that provides real–time access to ERP systems to employees and partners (such as suppliers and customers). The ERP II role expands traditional ERP resource optimization and transaction processing. Rather than just manage buying, selling, etc.—ERP II leverages information in the resources under its management to help the enterprise collaborate with other enterprises.^[17] ERP II is more flexible than the first generation ERP. Rather than confine ERP system capabilities within the organization, it goes beyond the corporate walls to interact with other systems. Enterprise application suite is an alternate name for such systems. ERP II systems are typically used to enable collaborative initiatives such as supply chain management (SCM), customer relationship management (CRM) and business intelligence (BI) among business partner organizations through the use of various electronic business technologies.^[18][19] “, Wikipedia

The definition held several false assumptions. I argued these to no avail with my fellow analysts:

• False Assumption #1. Transactional Efficiency is Sufficient to Drive Value. No doubt about it, ERP architectures improve the efficiency of order-to-cash and procure-to-pay transactional sets. We would not have built global supply chains without these capabilities. The investment in ERP sets the stage for an efficient supply chain, but misses the need for buffers, decoupling points, and policy shifts to improve agility. Excellence in supply chain planning requires constraint visibility and flow modeling to maximize value-based outcomes as we move from a focus on math to physics to understand supply chain dynamics. (For a better understanding, see the definitions below.)
• False Assumption #2. Tight Integration of SCM with ERP Improves Value. Tight integration of transactional data with planning models introduces nervousness and increases the bullwhip effect. (Don’t believe me? Measure it.) While ERP architectures are essential as a system of record, tight integration with planning limits the ability planning capabilities to model scenarios like what-if analysis and discrete-event simulation.
• False Assumption #3. The Integrated Supply Chain Improves Outcomes. While business leaders speak of the need for the integrated supply chain, the need state is interoperability. Tight integration without a unified data model and semantic reconciliation is a problem to drive data understanding and use cases like substitution, phase in/phase out of products, bill of material translation from PLM to SCM, and clarity on the selling unit.
• False Assumption #4. Enterprise Data Is Sufficient to Drive Excellence in Supply Chain Planning. The issues are many. In mature supply chain planning deployments, enterprise transactional data only represents 40-60% of the requirements. The reason? Enterprise data is not sufficient in three areas.
  • Newness. Brands grow through excitement and newness — creating excitement through new markets, products, and platforms. Enterprise data is backward-looking. Modeling newness in ERP-centric architectures is an issue. There is no good way to account for newness in tight integration of supply chain planning to ERP.
  • Market Data. When compared with market data (channel and supplier data), enterprise data has a latency of weeks or months. As a result, there is no place to put market data in ERP-centric architectures. Organizations that only use ERP data are always on the back foot seeing market shifts weeks or even months before reacting.
  • Network Data. The focus on enterprise-centric architectures was a barrier to building networks. By definition, ERP is not a good building block — it is linear and not bi-directional, and it lacks a many-to-many canonical — to define network flow.

Mistakes Made Over and Over Again

The market moves through shifts and hype cycles, repeating the same mistakes. In the 2000-2010 decade, the focus was on tight integration with ERP. Early in the next decade, the focus was on big data (use of unstructured and streaming data). There was then a shift to digital transformation, and now to AI. Over the last two decades, we have made little progress in supply chain planning.

Teams get caught in a hype cycle — failing to define terms and outcomes — falling prey to buzzword bingo. Alas, pretty words and pictures do not drive successful projects. Chasing shiny objects is a recipe for failure. Clarity on business objectives, definitions, and outcomes is essential. For example, today, AI is everywhere but nowhere in planning architectures. There are examples of agents and early forms of agentics, but the use cases are primarily to reduce planning workload, not to redefine work. The use of machine learning, reinforcement, and deep learning techniques is core to planning, but with all planning implementations, companies need to test and verify outcomes by backcasting, Forecast Value Added Analysis, Calculation of the Bullwhip, and testing plan feasibility.

The larger the company, the greater the gap in using insights to make decisions. As shown in Figure 1, consultants and technologists recognize the gap between data and decision-making, but few are working to close this gap. Instead, companies continue to chase shiny objects.

As a result, larger multi-nationals with a strong dependency on ERP-centric architectures face:

• Dependency on Spreadsheets
• Latency in Getting Data to Make Decisions
• Gaps Between Consultants/Technologists and Business Leaders/Information Technology (IT)

Frustration abounds. The pace of technological change in our personal lives, driven by digital transformation and artificial intelligence, is fast and furious. As hard as business leaders try, this shift cannot be duplicated in enterprise applications. The gaps are growing. The issue? Foundational architectures are a barrier to redefining work to unleash new levels of value. Just like a house cannot be built on a bad foundation, our current definitions of planning architectures need to change to take advantage of new capabilities.

In a recent research project, I asked, “What can we learn from digital transformation that we can apply to the testing of artificial intelligence projects?” In the research, we find that the most successful initiatives had the following characteristics (correlation at a 80% confidence-level):

• A business case using clear definitions with a focus on outcomes (absence of vagaries, hype, and shiny objects).
• Reporting to the line of business leaders (as opposed to IT leaders).
• Strong understanding of systems and design thinking.
• Inclusion of data scientists and supply chain finance talent on project teams.
• Test and learn mindset (versus a tight project plan based on a Request for Proposal (RFP)).
• Naturally curious: Questioning of historic architectures.
• Very little dependency on classical implementation consulting

These projects avoid vague terms like control tower, digital twin, continous planning and real-time decision making.

Learning from the Past

Last night, Jon from Clorox posted on LinkedIn about the need for continuous planning: a concept heard at an analyst conference. Thirty-eight people commented.

Similarly, at a conference last week, a business leader asked me for recommendations on real-time planning solutions.

This type of half-cocked discussion is not helpful. Planning data is never real-time. Instead, in a good implementation, data is available at the speed of business with zero latency.

Successful planning requires discipline and governance. The goal is to match the required data-acquisition speed to the definition of the planning horizon. By definition, tactical planning is a plan defined outside lead time (usually as part of Sales and Operations Planning). Lead time should be the sum of the source, transformation, and logistics lead times. It will vary by item and should be updated with each batch job using a planning master data layer.

Real-time data or continuous planning would only be a fit for the transactional planning horizon and should interface with Available-to-Promise and Allocation logic. This is often termed “sensing.” (Definitions matter.)

If companies do not adhere to manufacturing freeze durations in the operational horizon, noise and waste will result. Prove it to yourself. Measure outcomes.

Figure 2. Planning Horizon Definitions

If you go to a conference and you hear a new term or concept, clearly define it and test it before advocating deployment. (I tested the concepts of outside-in planning for three years and wrote on my insights.)

Wrap-up

There is no clear definition in the market for terms like control towers, digital twins, real-time planning, or continuous planning. And, just like I fought and lost the battle for ERP II as a Gartner analyst, you will need to fight within your organization to drive clarity.

My advice is simple. Take responsibility to train yourself.

Sidestep the hype, avoid word salad, get clear on outcomes, test and learn before deploying, and train teams on the principles of planning. Clearly define what makes a good plan and who should make it when. Measure planning effectiveness and evolve.

I would love to hear your thoughts. And, sorry, Jon, I don’t think the answer is continuous planning, despite the analyst group’s advice.

Definitions:

A key to understanding this article is the understanding of key terms:

• Transactional Efficiency: Transactional efficiency refers to the speed, accuracy, and cost-effectiveness of supply chain processes, aimed at enhancing business transactions and improving return on investment (ROI).
• Supply Chain Physics: Measurable mathematical relationships predicting the performance limits of the supply chain and how it will respond to both management action and external factors. The management of the supply chain requires recognizing constraints and designing three buffers: capacity, inventory, and time, with a focus on reliability. Supply chain physics concepts underpin the Theory of Constraints, which is foundational to supply chain planning.
• Supply Chain Dynamics: Supply chains are stochastic, dynamic systems driven by variability and shaped by complexity. The focus is on reducing variability, managing the rhythms and cycles of flow, and minimizing the bullwhip—to drive value. Supply chain dynamics concepts are critical to reducing bullwhip and driving effective reinforcement learning.

As the weeks move swiftly by, I feel each week that the industry is treading water. I see a lot of pretty PowerPoint presentations, but little innovation.

Technologists and business leaders are good at locking horns, but not dancing together to a new sheet of music.

Let’s Have Some Fun

This post is packed with a bit of musings, some sarcasm, and fun. Bear with me.

As technology advances, supply chain leaders are still fixated on automating planning processes designed in the early 1980s, when I still carried a slide rule in my briefcase. (Mine was round and quite cool, I might add.) Most business leaders approach supply chain planning as a math problem. It is so much more complicated.

In post after post, I see thought leaders writing about making today’s planning processes faster and more autonomous. My question is, “Should we really try to make these processes faster and more autonomous if these traditional processes are not effective in a more variable world?” I don’t think so.

My view? In this world of increasing variability, traditional supply chain planning processes are largely a cost sinkhole that adds little value beyond providing a company with a clear system of record (marching orders for the company’s future plan). In other words, the plan may not be good, but the organization has a common plan.

Why do I say this? As I test forecastability, inventory plans, plant schedule compliance, the use of market data for lead times, and the impact on the bullwhip effect, I am convinced that we are spending a lot of money to create waste. Today’s planning processes were a good fit for the small, regional teams, but, like the round sliderule, they are outdated for the global multinational. This is why we have focused on helping companies understand how to build outside-in processes.

You may not like this model or have your own, but it is clear to me that traditional Advanced Planning is only a fit for a regional, process-based company.

So, how do we move forward? I think that we innovate.

If we could adopt an innovation mindset rather than an improvement mindset, we could drive so much more progress. But to do this, we have to face the fact that performance is regressing across industries. With the widening of product portfolios and unchecked product complexity, items became less forecastable, making most traditional planning models archaic and increasing the bullwhip by 7-10X (based on homework from my outside-in classes). Few companies can measure and define a good plan.

Why Progress is Tough

The wheels on the bus go round and round. It is a world of lift-and-shift. The BlueYonder ex-employees’ bus went to Kinaxis; the Kinaxis employees’ train moved on to Relex; the Coupa teams moved to Optilogic and Lyric Software; and the o9 sales team moved to SAP. Supply chain planning is a story of re-treads. The world of supply chain planning is small, and many — especially in sales —hop from company to company.

In the process, the sales teams’ messaging and understanding change little. No technology or consulting company in the industry is good at measuring and delivering value. In the process, fresh thinking is extinguished.

(The moves from company to company are across the board. I just picked a few companies here. The names of the sales team personnel remain the same, but only the logo on their business cards changes.)

So, amid all the rah-rah speeches at technology kick-offs, we have companies full of nomads who are convinced that historical practices are best practices. Risk-averse and chasing deals, innovation at scale is largely non-existent. This is despite the confluence of promising new technologies. The promise has never been so compelling, and the deployments so disappointing in this old gal’s life.

Dancing With True Innovators

It is hard for a re-tread to dance with innovators, and the true innovators are far-and-few between, often willing to roll their own. To better understand the story, let’s examine some data.

Thirteen years ago, 22% of respondents in a study on S&OP technology stated they would be innovators, compared with 7% in a recent survey on digital transformation/artificial intelligence. Now, an astute reader will respond, “Yes, but are the samples comparable?” The answer is no. I don’t have an absolute number for innovators, but I do know that the percentage of manufacturers/retailers that see themselves as innovators ranges from 5-28% in my studies, with recent studies showing lower levels of companies identifying as innovators. I also know from the research that smaller companies are more likely to rate themselves as innovators. The global multinational is more conservative. They are also more confused.

Figure 1. Supply Chain Planning Innovators

More troubling is the ability of business/IT/technologists to work together to drive innovation. Look at the gaps in Figure 2. Manufacturers see themselves as significantly more effective than consulting partners, and software providers view them with 80% confidence when testing new approaches.

Sadly, industry acumen and understanding regressed over the last two decades. As a result, re-tread sales teams are forced to dance a nonsensical dance with late adopters. They both talk about innovation, but true innovation remains elusive.

Let’s face it, we don’t have good ways to test and learn on the potential of new approaches, which is why I am pushing technologists to push market education and Foundry models. (A business model where business leaders could come and test potential solutions to hairy problems before buying software.)

Figure 2. Ability to Test New Concepts

ChatGPT defines a foundry sales model as:

“Foundry sales models in testing software typically refer to frameworks or methodologies used to evaluate and deploy AI models effectively within software applications. These models help organizations assess performance, scalability, and integration capabilities of AI solutions in their testing environments.” ChatGPT

The test-and-learn model would allow business leaders and technologists to learn together, but it will require us to rewire the brains and bonus incentives of the sales teams that circle the market, shifting from company to company every year. It will also require us to do a Mea Culpa and admit that what was conceived in the period when I carted around a round slide rule is not a good fit for most organizations. Tough change management.

The technologists should take the lead, but few are capable of doing so. Business leaders should improve their capabilities to drive innovation and partner, but most are chasing shiny objects — talking innovation but unable to understand how to close the gaps. So, the retreads try to lead the dance, but make little progress.

How To Break the Cycle

My recommendation is simple. As a business leader, break the cycle.

Dos:

• Clarify business opportunities and align the organization on the definition of what makes a good plan. Remember that a good plan is actionable, feasible, and available at the speed of business.
• Approach the project with clarity, grounded in a systems approach and tied to strategy. Supply chain excellence is more than a math problem. Focus on improving balance sheet outcomes and aligning the organization from functional metrics to a balanced scorecard.
• Explore the use of market data and define process flows from the customer’s customer to the supplier’s supplier. Start by tackling the hairy problems that matter most.
• Redesign work based on discovery. Be open to the outcome.
• Build systems thinking in your teams. Look at the supply chain holistically.

Don’ts:

• Issue RFIs or RFPs.
• Hamper progress by building a project plan with a definitive ROI before the business case is clear.
• Tie the project to a fixed project plan before the objectives are clear.
• Staff the project with a bus-load of consultants. Remember, only one woman can have a baby. (The strongest talent is at boutique consultancies. Unfortunately, the consultants with household names are relatively weak in planning.

These are my thoughts. I welcome yours.

Fridays are my writing day. Each Friday morning, I make my pot of black coffee, put on my headphones, and begin writing. This morning, I had the opportunity to discuss the implementation of SCOR orchestrate with Peter Bolstorff, formerly at ASCM and now an independent consultant. He was a primary contributor to the SCOR DS model.

For those that have not followed the development of the SCOR DS model, the Level 1 processes are now orchestrate, plan, transform, order, fulfill, source, and return. Peter had three hours of road time, so we reviewed the model in detail. (Peter attributes his training on outside-in processes through Project Zebra as a defining moment to build the model in this manner.) We debated the term orchestrate in great detail. (You can imagine.)

We agreed that orchestrate was not a technology feature. Instead, it is a core organizational capability. (I wish that all the supply chain technology vendors leading with the term orchestrate would find all the references in their marketing copy and push the delete button.)

He ended the call by sharing a story on driving supply chain transformation using SCOR DS. As a strong believer in the Supply Chains to Admire methodology, Peter began this client engagement with metrics that drive market capitalization: growth, operating margin, inventory turns, and Return on Invested Capital (ROIC). After building orbit charts and analyzing market potential, he facilitated a session with his client to analyze organizational failure modes for 2025 performance.

His next step was to apply the SCOR Level 1 metrics and conduct a similar comparison. His conclusion? The biggest opportunity for his client is orchestrate. His recommendation to his client? Design a supply chain engineer role.

BINGO! I love it when great minds think alike.

Ganesh meet Peter. Peter meet Ganesh. Great minds are thinking alike.

Defining a Supply Chain Engineering Organization

I am a chemical engineer by background. Most academic programs offer materials, mechanical, computer, chemical, industrial, and electrical engineering, but I can find no school offering a Supply Chain Engineering degree. (Source Chat GPT.)

Many supply chain programs are housed in business schools and report through the marketing departments. (Which I think is amusing since supply chain management and marketing have such large gaps in alignment in the real world.)

I also find that Gartner’s analysis of supply chain management programs to not align with my personal experience. I struggle to find a good standard for measuring supply chain academic programs, and most of my manufacturing and retail clients frequently comment that students are ill-prepared to enter mainstream supply chain organizations. The concepts of a Supply Chain Engineering program appeal to me.

If you are a technologist, don’t fool yourself. Orchestrate is very different than concurrent planning. A technologist can enable orchestrate, but a business-led team needs to make it happen.

So, I went back to ChatGPT and asked for a job description for a Supply Chain Engineering role. Below, I share the output. I have crossed out the elements that I do not think fit and placed my comments in bold. I am interested in your thoughts. Here is a definition of orchestrate from the SCOR DS work.

Courtesy of ChatGPT: A Job Description

A supply chain engineer is responsible for analyzing and improving a company’s supply chain processes to drive shareholder value. enhance efficiency and productivity. In this role, they design and build supply chain orchestrate capabilities. They develop strategies, collaborate with various teams, and implement solutions to optimize operations and reduce costs value chains by improving results on the balanced scorecard of operating margin, growth, inventory, and Return on Invested Capital.

A Supply Chain Engineer is responsible for designing, optimizing, and improving the efficiency of supply chain processes within an organization. They analyze data, identify areas for improvement, and implement strategies to enhance overall performance.

Key Responsibilities

• Managing Complexity. Analysis of product lifecycle and platform options, design of new product launch capabilities, and management of end-of-life processes.  
• Design of Make, Source & Deliver Capabilities. Decisions on outsourcing, asset strategies, and time (order policy and cycles, manufacturing and supplier lead times, and route to markets) are made to strike the right balance between risk and opportunity. Risk should never be viewed in isolation.
• Definition and Management of Policy and Governance. Definition of contracts, strategic partnerships, supplier development, planning governance, and go-to-market strategies.
• Organizational Design and Building Supply Chain Core Capabilities/Knowledge Systems. Building supply chain acumen and collaborative systems. and workflow. Creator of career paths, mentoring systems, and development programs.
• Driving Innovation and Integrity in Data and Analytics Systems. Design of analytics systems to provide meaningful insights at the speed of business while driving innovation programs to test and learn the power of new analytic approaches.
• Alignment of Metrics to Maximize Value. Definition of benchmark and core capabilities, and alignment of reward systems to maximize value.
• Ethics and Governance. Policy and enforcement of ethics and rules of engagement. Definition of governance for returns, circular economy, and corporate sustainability.
• Process Optimization: Analyze and improve supply chain processes to reduce costs and increase productivity.
• Data Analysis: Conduct data analysis to identify bottlenecks and develop solutions.
• Collaboration: Work with various teams, including Sales, Operations, and Customer Service, to implement supply chain strategies.
• KPI Development: Determine key performance indicators (KPIs) to measure supply chain efficiency.
• Documentation: Document procedures for supply planning and train team members.

Required Qualifications

• Education: Master’s Degree in Organizational Development, Systems Theory, Analytics, or Data Science. Bachelor’s degree in Supply Chain Management, Logistics, Chemical or Mechanical Engineering, or a related field.
• Experience: 10-15 years of leadership in building teams and driving supply chain excellence. Geographically aware. The ability to manage and lead diverse teams. elevant experience in supply chain management or engineering.
• Skills: Strong analytical, problem-solving, influence management, and communication skills. Ability to work collaboratively in a team environment.

Discussion

Some might say, “Isn’t this the role of the Chief Supply Chain Officer?” Or, “The supply chain center of excellence?” My reply would be, “Only 5% of companies actively design their supply chains, and when they do, the objective function is usually cost mitigation. We speak of moving away from functional metric reward systems, but it does not happen. We speak of managing complexity and lifecycles, but it does not happen.” The reason? Most organizations are focused on building stacks of efficient processes to improve functional metrics. Systems thinking and the management of the system as a whole entity is missing.

The charters of most Supply Chain Centers of Excellence are not aligned to orchestrate. The gap is that companies are unclear about what defines excellence. If the organization is mired in driving functional outcomes, the Centers of Excellence will never successfully drive an orchestrated agenda.

Wrap-up

A common belief is that supply chain management is a math problem to be solved through linear optimization to improve efficiency. We must move past cost and functional metrics.

In reality, value is created by focusing on physics and flow, applying systems thinking while actively designing organizational capabilities, network/supplier relationships, and asset strategies to support the key processes of order, fulfill, plan, transform, source, and return.

Many organizations are in the midst of a digital transformation or an analytics revamp using AI, natural language processing, and large language models. As shared in last week’s newsletter, success happens when programs are business-led, with strong analytics talent (data scientists and programmers) on the team, and a clear understanding of the organization’s mission and values, executed through holistic systems thinking.

To do this right, I agree with Ganesh and Peter that we need a Supply Chain Engineer. However, we must move beyond conventional thinking about efficient process optimization to unlock true value.

Would love your thoughts. ll the best on this cold and rainy weekend.

Rabbit ears on top of the TV helped us get more channels.

We loved the new world of black-and-white TV. The transition to the color TV experience, when I was in college, was fantastic. The evolution of viewing options — Netflix, Amazon streaming, cable TV — stretched beyond our limited imaginations. In my lifetime, we have redefined home entertainment.

Reflection

I entered the world of manufacturing during the era of black-and-white TV.

At that time, sales and operations planning was a manual process. We forecasted based on historical sales and met monthly as a team to develop a potential line schedule. Not a lot has changed. Despite the implementation of advanced planning systems (APS) and enterprise resource planning (ERP), the process remains largely manual.

The most significant improvement came from the introduction of the Excel Spreadsheet. (Funny, but sad. Right?) Today, 92% of organizations operate some supply chain planning system; yet, as shown in Figure 1, 94% are highly or somewhat dependent on spreadsheets. Sadly, for most organizations, advanced planning became a system of record.

How do I form these opinions? I have completed 17 qualitative and 12 quantitative studies on S&OP and written 10 research summaries, first for Gartner, then for AMR Research, and finally at Supply Chain Insights. Many of the insights are summarized in the Shaman’s Journal series.

Figure 1. Organizational Dependency on Spreadsheets

S&OP should align the organization while providing insights on how to maximize the business strategy within the tactical planning horizon (outside of supply lead time). The best processes balance the “S” with the “OP.” Unfortunately, many processes today are hijacked by corporate politics. They are not balanced. Functional metrics and bonus incentives derail most processes.

So, what if we stretch our minds and redefine Sales and Operations Planning (S&OP) beyond the current state? Let me use a weather metaphor as an example.

Stretch Your Imagination

I imagine a world for S&OP where agentic AI develops multiple spaghetti-planning models for the coming months based on playbooks. The planner, as an orchestrator, shares the modeling parameters (through writing prompts) with the system before the planning period, after which the system develops multiple plans based on knowns and unknowns. The spaghetti models enable easy visualization of the unknowns.

I first encountered the concept of using planning books (multiple what-if analyses based on assumptions) when I was working with Eli Lilly in 2004. At the time, the process was manual. I wrote about it as an AMR Research analyst, and inquiries spiked. The demand was high, but there were no good technology options for my clients.

The process focused on likely business outcomes. At the time, Eli Lilly was competing heavily against Nova Nordisk in the launch of new diabetic treatments. The combination of platform delivery options (pill, injection, etc), dosage, and regional requirements drove uncertainty. Regulation precluded postponement.

Traditional APS approaches are designed for a world where items are forecastable and uncertainty is low. Recently, as I have been working with native AI platforms, the concepts of planning books and spaghetti models have merged in my brain as a potential option for S&OP.

What are Spaghetti models? It is a technique used in weather forecasting to display multiple possible paths for a storm, most often used during hurricane season. Each line in a model represents a different forecast run based on varying data inputs, creating a tangled “spaghetti-like” appearance. The primary function of a spaghetti model is to convey the range of possible storm paths, emphasizing uncertainty rather than a single model. This approach provides a visual representation of the possible trajectories for a weather system, offering insights to improve preparedness.

If we use spaghetti models to forecast uncertainty in S&OP, the supply chain planning system is connected to market data. (A planning master data system is continuously fed information on plant schedule adherence, quality of conformance, conversion rates, change-overs, inbound transit times by lane, prices, and labor/equipment availability.) The system’s output within the tactical horizon would be a set of spaghetti models to help business leaders gauge and visualize uncertainty.

To drive a plan to a system of record, each plan carries a probability. The range of probabilities across all plans forms the effective frontier for operating in the upcoming period. As new data becomes available, the plans are consumed — from tactical to operational planning — based on market data. S&OP stakeholders can visualize plan consumption daily and post questions and concerns for the model.

Key to this new process is agreement by all stakeholders on “what good looks like.” This includes customer segmentation prioritization, a balanced scorecard (functional metrics are replaced by reliability goals, enabling the functions to support improvement of the balanced scorecard), and the management of complexity. For most, this will be the most challenging task.

Meetings are virtual. Goals are aligned. No one feels pressured to buy into THE plan. The multiple plans help teams visualize uncertainty. Each model has a probability, and the analysis of all the probabilities defines the effective frontier of outcomes. This frontier of outcomes helps to translate demand into planned orders and aggregate buying plans for commodities and direct materials/transportation requirements.

These models and probabilities are available for all to view. (Not only the plan, but the impact of the plan with the enablement of self-service modeling for what-if analysis.) As the calendar progresses, a small team guides plan consumption and, when there is a choice, answers modeling questions. (An S&OP execution team.

Wrap-up

What I see in demos from tech providers aiming to drive value through agents is automation of today’s processes. A better forecasting engine. Probabilistic modeling. A deeper inventory optimization approach. The better use of streaming data for sensing. Yadda yadda.

So, many experts are applying new technologies to improve current processes. Yawn.

I don’t see anyone attempting to redefine S&OP in the same way Netflix redefined entertainment. I feel that we are caught in the AI stupid lane, fiddling with rabbit ears on the black-and-white TV as the world of technology evolves around us.

As we seize opportunities and potentially drive innovation, I am thinking a lot about thinking styles and the alignment of critical, design, and systems thinking to capitalize on the potential of new approaches to drive process automation. The reason? We are stuck as an industry.

Note in Figure 3 that manufacturers rate themselves higher on thinking styles than consultants and technologists do. Teaching critical, design, and systems thinking is an opportunity for both internal training programs and academia.

The supply chain is a complex, non-linear system with rising uncertainty. Current processes model the relationship between knowns and knowns, but miss modeling unknowns.

Improving planning processes requires critical and systems thinking. Making new approaches usable requires design thinking.

Figure 3. Analysis of Thinking Styles

OK. Here I share a vision. I am sure that there will be a lot of naysayers, and that is ok. I have thick skin. My goal in this blog is not to provide AN ANSWER. My goal is to stimulate conversation and for us not to AI Stupid. As uncertainty increases, we need to look past traditional definitions of APS.

I look forward to hearing your vision. Please join the dialogue. Let’s break traditional norms because today’s S&OP processes don’t work so well.

Meanwhile, I am busy working on the analytics of the artificial intelligence study that I just pulled from the field. If you helped me and answered my survey, I send a hearty thanks!

Happy holidays to all my readers.

I remember sitting at SAP Insider listening to a compelling speech on digital transformation in the spring of 2017. As I listened, I scratched my head, and asked myself, “What does it mean to have a successful digital transformation?” What does digital mean in this context? And, how do we drive value?”

You know me. I am a geek. Definitions matter. I despise the use of hype-based rhetoric by technologists.

I then set off on a mission to gain more insights. The article I published was, “Are you active in the Digital Transformation Drinking Game?” My conference in 2018, showcased case studies of digital transformation. My favorite was the transformation of manufacturing by Agco, a manufacturer of heavy equipment. This one suprised me. It was brilliant. But, I found few to equal.

Since then, digital transformation is everywhere but nowhere. Lots of hand waving, and rhetoric. So, in my recent analytics survey, I wanted to find out in greater detail where companies feel that they got value. In short, manufacturers over 5B in annual turnover drove improvement in transactional efficiency in procure-to-pay and order-to-cash eleminating paper processes, but few drove value in supply chain planning or visibility.

Goal

The primary goal of digital transformation was to improve planning and decision making. The announcements by SAP on SAP APO and the movement to SAP IBP, triggered many companies to evaluate alternatives driving an increase in spending in supply chain planning. Within the organization, 55% of the programs were driven by the CIO, and more than 45% of the programs are more than four years old.

The digital transformation model built by Deloitte with ASCM added to the hype. In my opinion, It was just bad. (Feedback shared multiple times with both Deloitte and ASCM.)

Figure 1. Goals of Digital Transformation

The Results

When you look at the results, companies were more successful in improving transactional efficiency than improving planning and supply chain visibility. The redefinition of transportation and the use of transportation streaming data remains an opportunity that few believe is currently driving effectiveness. (The issue is that we have to redefine Transportation Planning to gain true value. This is subject for another blog.)

Figure 2. Self-reported Effectiveness

My Take

The results do not surprise me. During the period, I conducted many qualitative interviews. I found some consistent themes:

• IT Standardization Is a Death Knell for Planning. Sounds good. One throat to choke. A common vendor. However, planning is industry specific, and needs to be driven by business requirements. IT standardization as a mandate for supply chain planning seldom results in an effective deployment. And, deployment by a large consultancy is fraught with a high potential for failure.
• Many Companies Used Digital Transformation as a Foil to Spend More on ERP. With the rise in maintenance costs and the reduction in budgets, many CIOs used digital transformation as an excuse to spend more on unfinished projects.
• Few Companies Understand Supply Chain Planning. As I speak and work with clients, I am surprised at how few companies (this is includes most consulting firms) understand the basics of supply chain planning. The lack of clarity on requirements is the first of many stumbles when the functionality and work requirements are not clear.
• Delivery Methods Lacking. With the rise in spending, the deployment methods of the supply chain planning technologists were not equal to the challenge. Most assumed the client was clear on the requirements. Most were not.

This is my take. I would love to hear from you.