Most tools were never designed to work together in one operating environment.
“Two Decades of Tools, and the Systems Still Don’t Talk”
A new commentary on CSCMP’s Supply Chain Xchange gets at something I’ve been saying for years: supply chains “have become increasingly connected over the past two decades but simultaneously have become increasingly fragmented.” Twenty years of buying tools, and each one solved a real problem. But as the author puts it, most of them were “never designed to function as a single operating environment.” So, the more tools an operation bought, the more disconnected it got.
That sentence describes most of the freight operations I talk to. It also describes exactly what we set out to avoid when we built EKA, so let me lay out the case.
The Seams Show Up When It Matters Most
The CSCMP piece points out that fragmentation gets expensive the moment disruption occurs. When something breaks, disconnected systems slow communication, complicate decision-making, and force people to use manual workarounds to keep freight moving. And those workarounds don’t stay temporary. They harden into how the work gets done every day. In today’s fast-moving supply chains, every minute lost to manual coordination is money out the door. Sticking with outdated methods isn’t just inefficient—it’s a recipe for escalating costs and missed opportunities.
I watched it up close a few days ago. I sat with a 150-truck carrier, a customer, and before showing them anything, I asked how they plan their loads today. They work in three southwest markets. A planner looks at every truck coming into each of those markets, checks each one’s hours of service, then hunts for loads on the other side. He rebuilds that picture five times manually in 2026. These are smart people running a real fleet. Their tools just never talked to each other, so a person spends his day doing the connecting that the systems should have done.
Soon they will be using a proprietary EKA load-asset optimizer with a set of tools: visibility, HOS, predictive ETAs, on-time pickup and delivery monitoring and alerts, dwell time decision support, driver home time and other driver constraints, operating market area controls, etc, from EKA that are housed and driven by a single unified data layer. All designed to function in one operating environment, resulting in the highest productivity load planning and dispatch workflows, valuable insights and rapid decision-making
AI Made the Fragmentation Impossible to Ignore
For years, operations just lived with the fragmentation. AI is making that impossible. Gartner surveyed 140 senior supply chain leaders at organizations doing $250 million or more, and 56% named AI and legacy systems as a major challenge, the top roadblock on the list. The CSCMP author reads it the way I do: the weakness was always there, and AI is simply exposing it.
You can see it in how these projects actually go. A company buys an AI point solution with big promises, and onboarding stretches for months because the vendor discovers the data is fragmented, the workflows were never written down, and every record lives in a different system. Some vendors now station their own engineers inside customer operations just to stitch the plumbing together, and when a product needs that much field surgery to work, the problem was never the product. It’s the same reason so many AI agent projects fail: the agents are usually fine. The ground under them isn’t.
The newest wave of agentic tools repeats the pattern. An agent that sits on top of somebody else’s TMS inherits every seam underneath it: the data quality, the data fidelity, the gaps in what got captured in the first place. We wrote about where the baton gets dropped, and every one of those handoffs is a place where an agent loses the context it needs to make a good call.
One Environment
Here’s the fix, as plainly as I can put it: one operating environment. One database, so every tool works from the same record. Workflows that run where that record lives, instead of exporting to a second system and hoping the sync holds. Insights computed on data that’s current because it never left. Decisions are made rapidly and where the work happens.
A TMS built this way stops being one more tool on the pile. It becomes both the system of record and the system of execution, and that combination is what everything else depends on. Legacy and cloud-based platforms hold the record but can’t execute. Point solutions execute one task but don’t hold the record. Stitch the two together with integrations, and you’ve recreated the fragmentation the CSCMP piece describes; this market is testing it in real time.
Why We Built the New Agents Inside, Instead of on Top
This is why the eight agents launched year-to-date and four AI agents EKA launched just this week run natively inside one Omni-TMS environment: a control center that pushes data and event exceptions to the person who needs them, a trip fuel optimizer, automated handling of PDF and spreadsheet load orders, and a load and asset match optimizer. The count matters less than where they run. Because the agents live inside the environment, there’s no integration project and no months spent cleaning up data before they can start, since they run on the same record the rest of the operation already runs on. When the optimizer builds a plan, it already knows the driver’s hours, the dwell history, and the lane economics, because that data was never anywhere else.
Is that carrier planning three Southwest markets by hand five times a day? The goal is for that picture to build itself overnight, on the system they already run their freight on, so their planner can start the morning reviewing a plan instead of building one. Nothing gets bolted on, and nothing needs reconciling.
The Bottom Line
Most operations don’t need another tool. They’re already carrying too much, and the CSCMP piece describes the bill: slower communication, harder decisions, and people spending their day retyping what the systems should have shared on their own. So, before the next purchase, ask one question: Does this live in my environment, or does it add another seam? Ask it about every AI pitch that lands in your inbox this year, including ours. We built EKA to answer it well, with one environment, one record, and the intelligence running natively on both. Come see it.