... but becoming the biggest winner in a new era of logistics
By: Per Imer, CEO, Homerunner
Contains: 1075 words
When major technological shifts are discussed, B2B is often portrayed as the heavy train. Slow to accelerate. Careful by nature. A market that only moves once everything is proven, regulated, and standardized. Innovation is expected to happen elsewhere first, and B2B is assumed to follow later.
But in the shift toward Logistics as a Service, something unexpected is happening. The very characteristics that historically made B2B complex, cautious, and difficult to automate are now what position it as the biggest winner. Not despite complexity, but because of it.
There are moments when the world does not change by becoming slightly better, but by becoming something fundamentally different. These moments are rarely obvious at first. They are felt gradually, when what once held together through experience, commitment, and solid systems begins to strain. When organizations notice that maintaining the same level of stability suddenly requires far more effort.
Logistics has reached such a moment.B2B logistics did not follow the same trajectory. It remained manual, fragmented, and heavily dependent on people. It has been tempting to explain this gap by pointing to lower maturity, weaker investment, or lack of ambition. But that explanation is too simple.
B2B logistics was not slow. It was structurally difficult.
What we are seeing in the market today is therefore not just another wave of automation. It is a shift in how complexity itself can be handled.
For decades, logistics evolved by adding layers. More systems. More integrations. More processes. More people. Each time complexity increased, another layer was placed on top, and for a long time, this approach worked. Volumes grew, flows accelerated, and efficiency improved.
What has grown the most, however, is not the volume of goods. It is the volume of decisions.
Every order today represents a decision space. Price, delivery time, risk, customer commitments, exceptions, and downstream consequences that may only become visible later in the chain. Logistics has become a cognitive system, where value is increasingly created not in movement alone, but in the decisions that govern movement. Yet it is still often treated as if it were primarily an operational challenge.
Many organizations feel this pressure as a quiet fatigue. Not because their people lack competence. Quite the opposite. But because they are constantly compensating. Filling the gaps between systems. Handling exceptions. Absorbing consequences after the fact.
It shows up in manual corrections, in special handling, and in the familiar phrase, “This is something we usually manage.” The problem is that what usually works does not work in a world that changes while decisions are being made. As variability increases and the cost of error rises, experience alone becomes a fragile stabilizing force.
AI does not change logistics simply by making existing processes faster. It changes logistics by shifting the boundary of what can realistically be carried.
For the first time, systems can hold multiple perspectives open simultaneously. Price, time, risk, probability, and consequence are evaluated not sequentially, but continuously, as the world moves. AI does not judge. It does not form opinions. It weighs.
Decisions move from being reactive to becoming preventive. Deviations are detected while they are emerging. Logistics transitions from being monitored to being orchestrated. But how AI does this, and what it is even capable of seeing, depends on something more fundamental than the model itself.
It depends on the architecture.
There is something appealing about closed data structures. They promise order, clarity, and control. The world is reduced so the system can function. Decisions become fast, consistent, and locally correct.
But every time data is closed in, something else is excluded. What happens across systems. What emerges later. What does not fit neatly into a predefined field but is still real. Closed data reduces the world to make it manageable. At that point, intelligence begins to resemble efficiency without understanding. Open data is not chaos. It is an acknowledgment that the world cannot be reduced without loss, that decisions always have ripple effects, and that truth is rarely local.
Open data forces systems to see the whole, not just the event. It does not make decisions perfect, but it makes them meaningful.
There is a reason the story of Ultron continues to resonate. Marvels Ultron figure(*) is created by people who are tired of reacting too late, tired of knowing that the next threat will arrive faster than they can comprehend it. He is born from care and the desire to protect.
But Ultron receives a goal without a life. Data without relationships. He becomes dangerous not because he wants the wrong thing, but because he sees too little.
In logistics, Ultron is rarely dramatic. He lives in dashboards, rules, and automated decisions that appear correct in isolation and wrong in the larger context. The problem is not the algorithm. It is the slice of reality the algorithm was allowed to see.
Avoiding the Ultron effect in real systems requires a space between intention and action, between event and consequence, between efficiency and understanding. This is where a middleware layer and an API first mindset become essential, not to own data, but to connect it. A layer where events can flow freely across systems in real time.
When this layer is combined with real time data capabilities, something fundamental changes. Data is no longer history. It becomes presence. Decisions are made based on movement in the present, not yesterday’s snapshots.
This is not where logistics is automated, this is where it is orchestrated.
It is in this space that Logistics as a Service truly makes sense. LaaS is not software delivered as a service. It is responsibility delivered through open connections. When responsibility is taken for outcomes rather than systems, local optimization is no longer sufficient.
In complex B2B environments, where variation is high, consequences are significant, and blind optimization carries real cost, this shift is not theoretical. At Homerunner, we already work in practice with LaaS across such environments. We do not build another system. We build the connections between systems.
When open data, middleware, and real time orchestration converge, logistics begins to respond before humans escalate, and decisions start to reflect the whole rather than isolated events. Logistics is no longer merely the movement of goods. It is the movement of decisions. And the difference going forward will not be who has the smartest AI, but who has the architecture and the willingness to give intelligence enough of the world to see.
LaaS is not the future because it is clever. It is the future because it is necessary.
------
(*) Note: Who Ultron is (in short)
Ultron is a Marvel Comics supervillain and one of the franchise’s most iconic examples of runaway artificial intelligence. After rapidly evolving, it concludes that humans themselves are the greatest threat to peace—and decides they must be controlled or eliminated.
