Why Supply Chain Never Produced a Category-Defining Company

Every major enterprise category has a clear leader. Supply chain has dozens of tools and no winner. The reason is not technology. It is that the industry kept selling software when customers were only ever buying outcomes.

Name the category and the company comes to mind immediately. CRM has Salesforce. HR has Workday. IT service management has ServiceNow. Data has Snowflake.

Supply Chain has no one.

It is one of the largest and most consequential categories in enterprise software, sitting underneath every physical product in the economy, and it has never produced a defining company. It has produced dozens of good ones, each owning a slice: demand planning, supply planning, warehouse management, transportation management, yard, visibility, procurement. No single company owns the result.

I have spent eleven years inside this problem. I think I finally understand why.

Supply chain is a chain, not a function

CRM is a function. HR is a function. You can wrap a function in one system and win the category.

Supply chain is not a function. It is a sequence: plan, source, make, move, deliver. Each link grew its own software, its own vendor, its own data model, its own definition of success. A typical Fortune 2000 supply chain now runs on 10 to 15 separate tools, each optimizing its own link, none of them accountable for the whole.

So the category fragmented by design. Every vendor optimized locally and left the enterprise to assemble a result out of the pieces. The enterprise never got software that owned the outcome, because no software was built to.

The orchestration graveyard

The industry saw this gap and tried to fill it, repeatedly.

The attempts had names: control towers, orchestration layers, visibility platforms, single panes of glass. (FourKites helped build that last category, and I will come back to that.) Every one of them promised to sit above the point tools and tie them together.

Every one of them stopped at the same place: a dashboard.

A dashboard is a description of the problem. It shows you that a shipment is late, that a lane is underperforming, that inventory is in the wrong place. It does not fix any of it. It hands the work back to a human and asks them to go act in the ten other tools. We did not unify supply chain. We added one more screen to look at.

That is the quiet failure at the center of this category. Decades of software, and the hardest part of the job, actually resolving the exception and actually moving the number, still landed on a person.

Nobody was ever buying software

Here is the reframe that changed how I think about our own company.

No VP of supply chain wakes up wanting to own more software. They are measured on whether product arrives on time and in full, at the right cost to serve, without burning the budget on expedited freight. They are accountable for a number. Software was only ever a proxy for that number, and a weak one, because it still required them to do the work of turning a tool into a result.

We sold capability and called it value. The customer had to supply the value themselves.

The real reason rip-and-replace fails

This also explains something every enterprise software leader has watched happen and quietly blamed on the customer.

When a company switches from one supply chain vendor to another, the project often struggles. The reflex is to blame the losing vendor's product. In my experience that is usually wrong. The new product is frequently better. The thing that breaks is change management.

A traditional tool asks people to change how they work before it gives them anything in return. The burden is upfront and the value is deferred. You are asking a team that already has a hard job to learn a new system on top of it, on the promise that results will come later. Most of the time, that is the real reason transformation stalls. Not the technology. The human cost of adoption.

You cannot win a category by asking people to work harder before they see a result.

Meet users where they are, by delivering the outcome

The way through change management is not better training or a smoother onboarding flow. It is to stop charging adoption as the price of value.

If an AI agent does the work, resolves the exception, chases the carrier, reschedules the appointment, protects the on-time-in-full number, then the customer gets the result without first reorganizing their team around a new screen. Value comes first. Adoption follows, because it is earned rather than demanded. That inverts the change-management problem the industry has been losing to for decades.

This is what becomes possible now and was not possible before. For most of the history of this category, the only thing software could do was show you the work. Now it can do the work. When the unit of value shifts from the tool to the completed task, the outcome itself becomes the product.

We call it Outcome-as-a-Service for Supply Chain™. Every engagement starts with the number the customer is on the hook for, and we work backward to the agents, workflows and data required to move it.

Why this is defensible, and why now

An obvious objection: if AI agents can execute the work, what stops anyone from doing this?

The models will not. AI models are becoming a commodity; any serious vendor will have capable ones within a release cycle of each other. What cannot be copied is the real-world context an agent needs to act correctly on a customer's behalf. Whether an outcome is met is usually decided outside the customer's own four walls: how a carrier actually performs on a lane, what is happening at a port, what the receiving dock looks like right now. That signal does not live in any one company's systems.

It lives in a network. Ours connects more than 1,600 brands and 500,000 carriers across 176 countries, built over eleven years. Every exception an agent resolves feeds back into that network and sharpens every outcome built on it. Commodity models plus proprietary network context is what lets us commit to a result instead of just reporting on one.

That is also why our mission changed. For a decade it was about building the network. Now it is about the work the network completes: one billion hours of operational work done for our customers over the next decade. A mission you can count, measured against a number the customer can feel.

The company that wins supply chain

The defining company in this category was never going to be the one with the best dashboard. It will be the one willing to take responsibility for the outcome, and able to back that commitment with data no one else has.

That is the company we are building. We are relaunching as fourkites.ai, organized not around a catalog of products but around the outcomes our customers are measured on, with an agentic advisor named Owen that opens every conversation with your problem rather than our software.

Supply chain has waited a long time for someone to own the result. We intend to be that company.

Curious what this looks like for your supply chain? Talk to an Outcome Advisor.

‍