A mid-market shipper with $200M in revenue often runs their supply chain better than a $10B enterprise. Fewer systems, fewer people, one language for how work gets done. The enterprise has real advantages, scale, talent, capital, but coordination is not one of them. That is why orchestration is a harder problem at the top of the market than the middle.
The systems problem
An enterprise shipper does not run "an ERP." They run three, a legacy SAP instance for the North American plants, an Oracle E-Business Suite in the European subsidiary, and a JD Edwards installation that came with an acquisition four years ago. Add two TMS platforms, a WMS in every DC, and a mix of carrier portals, and no single system holds the truth about a purchase order.
Every orchestration effort collides with this reality. The instinct is to consolidate, one ERP, one TMS, one WMS, but the transformation program is three years and a nine-figure investment. Nobody wants to fund the coordination work that has to happen in the meantime.
The pragmatic answer is a platform that sits on top of the existing systems and normalizes their data into one operational record without forcing a re-platform. Our comparison of ERP extensions versus digital SC platforms covers why this is the right architectural choice for large environments.
The org chart problem
Orchestration is fundamentally cross-functional. A single complex shipment involves procurement (raised the PO), the supplier (built and shipped it), logistics (moved it), operations (needs it on the line), finance (paid for the expedite), and quality (approved the substitute part). At an enterprise, each of those functions has its own director, its own tools, and its own KPIs.
When something goes wrong, ownership fragments. The procurement team says the supplier missed a ship date. The supplier says logistics booked the wrong lane. Logistics says the carrier changed the ETA. Everyone is technically right. Nobody is accountable.
The fix is not more meetings. It is a shared record: one case object per shipment, with a single named owner, a decision log, and a cost tally, visible to every function that has a stake. The control tower model is one version of this, though most tools stop at monitoring and never get to the ownership layer.
The data problem
At mid-market scale, missing data is a nuisance. At enterprise scale, it is the default state. Suppliers send inconsistent EDI. Carrier ETAs update on different schedules. Customs status arrives in a broker email nobody reads until Monday. The result is a data set that would embarrass the analytics team if it were surfaced, so it is not surfaced. Decisions get made on gut feel and Slack threads.
The way out is to accept that the data will never be perfect and design the platform around handling conflict, gaps, and stale values gracefully. Our post on how real-time data helps with decision-making in supply chain covers what "good enough" looks like in practice.
The change management problem
Enterprise supply chain teams have been through more transformation initiatives than any other function in the business. Every new platform arrives to a workforce that has been told, five times before, that this one is different. Skepticism is rational.
The teams that succeed with orchestration do not try to change everything. They pick one lane, one plant, or one customer segment, instrument it end to end, prove the outcome in a quarter, and let the story spread inside the business. That is a slower ramp than the vendor slide deck suggests, and a much higher hit rate.
Where to start
Pick a single lane where the failure cost is high and the current process is visibly broken. Instrument every case for a quarter. Measure MTTR, cost of expedites, and root cause distribution. Do not build to the whole enterprise on day one. Do not try to integrate every system before shipping value. Ship a first defensible win, then earn the right to expand.
How Orkestra fits
Orkestra was built for exactly this shape of problem, a large operational footprint sitting on top of many systems, with cross-functional ownership and imperfect data. The platform connects to what you already run and delivers real-time visibility, exception management, analytics, and AI agents in one operational view, so every shipment, order, and cost stream lives on one shared record. Customers deploy the first lane in 12 weeks, prove the model, and expand from there. If you are stuck at the orchestration ceiling, book a workshop and we will walk through one of your lanes.
If one of those systems is an SAP instance in the middle of an S/4HANA migration, the move is the same: run a focused supply chain project alongside the migration rather than parking it until go-live.
