I should have known better. I work in supply chain. Even worse, I work for a company who makes solutions to help companies solve these kinds of problems, which I listen, speak and write about for a living! What was the problem? I ordered a table online without checking reviews sufficiently. If I had done due diligence, I’d have seen the warnings – the retailer had pretty pictures on their website, but their supply chain failed them, creating an eleven-month saga and unhappy customer. The combinatorial factors disrupting supply chains persist, but in this case the problem was completely preventable. The issue wasn’t poor planning – they had the inventory. The failure was in supply chain orchestration – connecting planning to execution.
Supply chain orchestration failure – a table fable
What happened? In February 2022 I ordered a lovely midcentury modern table. When the 3PL driver delivered it a few months later he noticed it was damaged and submitted a request for a replacement. Months went by, but we were in the thick of the pandemic, with global supply chains under immense pressure, so I just assumed materials shortages were to blame.
By the end of the year, I realized that the arrival window was merely being postponed automatically and repeatedly at the same cadence, so I called customer service. Multiple calls only muddied the waters, but a few things became clear: inventory was in the warehouse, but my order for it was stuck. The distribution center (DC) hadn’t released the order, but customer service didn’t have access to the right systems to see exactly what was wrong.
Only persistent calls got my order back on track. Customer service couldn’t call the DC, only email them, and her emails weren’t getting responses. The order management system wasn’t effectively connected to the warehouse management system showing tables available or the transportation management system showing the carrier had returned my first table. The “swivel chair” effect was in place – information about my order required customer service to switch between multiple, disconnected applications, if they even had access. A failed process, a lack of tactics to resolve that failure.
What are the lessons to be learned from this debacle? First, we need true supply chain orchestration, connecting the plan to the execution. Second, automation needs to keep humans in the loop for monitoring. And third, supply chains must be designed to scale to ensure customer satisfaction.
First lesson: supply chain orchestration must connect planning to execution
Supply chain orchestration is about ensuring success from the first mile to the last. As supply chains have grown in complexity and length, siloed thinking and processes are no longer able to meet this objective. And as much as AI is at the top of the hype cycle right now, AI wouldn’t have solved this problem. Even if this retailer had used AI to forecast my purchase, AI that creates a perfect plan but fails to execute it only results in a highly-efficient silo, as Northeastern University professor Nada Sanders says. Planning was effective in this case, execution failed. Supply chains are “keepers of the last touch,” essential for a customer experience that delights…or disappoints. Orchestration built on legacy, disconnected systems may keep costs down but will not delight customers and grow top line revenue.
A supply chain should be set up for the end-to-end orchestration of people, and physical, digital, and financial assets, to meet customer expectations and corporate goals. This orchestration depends on three flows – information, physical, and financial – but when functions are siloed flows are blocked or delayed.
Planning is responsible for forecasting how many people like me would want to buy this table, starting a flow of information ensuring materials were ordered, and production and delivery scheduled. Once a return was triggered in the system, information flows failed. The company presumably had a process in place, but what appeared lacking were tactics for dealing with the inevitable disruptions that cause process failure. They delivered a table to me initially, but when it was returned due to damage, something in their process failed, the information flow stagnated. Events and disruptions are par for the course in supply chain, so we need tactics to kick in for problem resolution. Their tactics were not able to route my table correctly in their order management system once the return process was initiated.
Second lesson: automation needs to keep a human in the loop
The second lesson is that automation should augment humans, not replace them. Supply chain leaders from planning to procurement to logistics face labor shortages and cost concerns, so they want to reduce workforce time spent on manual activities. Automation can increase productivity by eliminating many tedious, repetitive tasks. But as we augment humans with automation we shouldn’t replace them. Automation, especially if it includes AI, needs to keep a human in the loop.
While analysis can suggest parameters, similar to upper and lower control limits, planners should ultimately set the parameters, based on their knowledge of the business. When there’s a gap between designed and actual results outside of these limits, alerts can trigger a human to investigate the exception. My table order was updated automatically for months, and likely generated an exception message, but humans in this kind of loop feel “alert fatigue” induced by thousands of exception messages. Another failure in tactics.
Third lesson: design supply chains to scale and satisfy
Finally, supply chains must be designed for scale. It’s not enough to make a model table for a website or a showroom. The table needs to be able to be manufactured and delivered to meet demand on time in full (OTIF), to achieve customer satisfaction. Eleven months later, when my replacement table finally arrived, it was again damaged in delivery, a pattern I found repeated in other customer reviews. I declined their offer to send yet another replacement, because I was too irritated to put up with them again. Is that the kind of customer experience they wanted?
Something is wrong in this retailer’s process, whether it be materials used, packing, shipping, etc., that made OTIF fail repeatedly at scale. Costs saved in production, packaging, distribution may have delivered results these functional areas are measured on, but this siloed approach suboptimized the ultimate objective, a happy customer. Supply chains can make or break a customer experience and should be seen as an asset, not merely a cost center.
Bonus lesson: remember that you work in supply chain
There’s also a parallel bonus lesson here for supply chain leaders. Building capabilities for supply chain orchestration, automation, and scale requires modernizing legacy, disconnected systems. And employees to run them. According to research by the boom! Global Network, 48% of the supply chain professionals in their 2023 survey on Next Generation Talent did not feel the technology at their company helps them do their job efficiently and effectively. And 39% felt the supply chain systems and technology at their company fell below their expectations. Will this environment attract the next generation of talent?
When investing in a modern supply chain software solution, remember that you work in supply chain. Check the reviews of your provider. Make sure that their end-to-end orchestration results in your promised solution can be delivered on time in full. Even if their solution demos impressively, look past the shininess for proof ensuring your solution will go live in the time frame you expect. Do they promise you a magical, autonomous supply chain powered by AI? AI works best when it is human-centered, so automation augments our capabilities. And can the solution scale beyond a proof-of-concept across all teams, business units, geographies, without perpetual services? Ask the questions. I wish I had.
Polly Mitchell-Guthrie is the VP of Industry Outreach and Thought Leadership at Kinaxis, the leader in empowering people to make confident supply chain decisions. Previously she served in roles as director of Analytical Consulting Services at the University of North Carolina Health Care System, senior manager of the Advanced Analytics Customer Liaison Group in SAS’ Research and Development Division, and Director of the SAS Global Academic Program.
Mitchell-Guthrie has an MBA from the Kenan-Flagler Business School of the University of North Carolina at Chapel Hill, where she also received her BA in political science as a Morehead Scholar. She has been active in many roles within INFORMS (the Institute for Operations Research and Management Sciences), including serving as the chair and vice chair of the Analytics Certification Board and secretary of the Analytics Society.