Mark Wallin, GM and SVP of Product at Phillips Connect, drives the future of freight through AI, automation and connected trailer tech.

getty

The conversation in transportation right now is dominated by a single question: When will autonomous trucking arrive at scale? It is a legitimate question, but for most of the operators I talk to, it is probably not the most useful one to be building decisions around. The more pressing question is whether their operations are actually built to absorb that shift when it comes.

Autonomy and automation get used interchangeably, but they are not the same problem. Autonomous trucking involves a fundamental change to the role of the driver, and while commercial launches are underway on select corridors, broad deployment across most routes is still years away. Automation is something leaders can act on today: removing friction from work that already exists—from manual verification processes that connected technology should handle to maintenance schedules driven by the calendar rather than equipment condition to service failures that surface only after a customer calls.

What connects them matters because autonomous systems, when they eventually scale, will require the same foundations that good automation already demands: clean data, consistent processes and clarity about how decisions get made. Organizations that build those now will be ready to absorb what is coming. Those that skip it will find that removing the driver does not remove the underlying complexity; it just makes it harder to ignore.

In my experience, the most meaningful gains rarely come from sophisticated technology. They come from fixing the places where people work around systems instead of through them. This is the pattern I keep seeing: An organization invests in a new platform, runs it for a year and sees nothing change. When you ask what they did differently, the answer is usually not much. In one case, driver performance data was generated every week, but no one had built a process around what to do with it. Leadership knew who the lowest performers were and had no system for acting on it, and the investment delivered nothing because the operation never changed around it.

The 2025 Gartner Future of Logistics Survey found that "94% of supply chain logistics leaders either currently have a supply chain management transportation system or plan to deploy one in the next one or two years." Yet in that same survey, 40% noted "the struggle to realize value from existing technology investments as one of their top 3 challenges."

The real competitive separation I see developing is not between organizations that have invested in platforms and those that have not; rather, it is between those that have changed how their people work and those that have layered new software on top of the same old processes. In transportation, you can see it quickly: In one operation, current information reaches the right person, and a decision gets made; in another, the platform is running, but the team still operates from habit, using data nobody trusts. The difference is whether the operation actually runs the way it should.

What the organizations that get the most from automation have in common is that they start with one problem, not everything at once. They identify a specific constraint—whether that is turnaround time, equipment availability or a maintenance backlog that keeps producing surprises—and focus there first. The easiest way to find where to start is to look for the workarounds: a team member maintaining a personal spreadsheet because the shared system is not trusted, a manager who learns about a problem by phone instead of an alert. Those gaps are a direct signal of where automation will return value fastest.

Before automating how exceptions get handled, agree on how they should be handled. Automation makes a process faster but not better. If an equipment issue creates disagreement about who responds, automating the alert produces a faster version of the same confusion.

The metric that surprises most leaders when they first track it is decision latency: the time between when an alert fires and when someone actually acts on it. Closing that gap through better workflow design frequently delivers more impact than adding another data source would.

Fragmented processes and siloed systems do not disappear when a vehicle no longer needs a driver. They get exposed. I have also watched organizations make the mistake of automating the wrong thing entirely. A fleet that has long required manual pre-trip inspections may digitize the form without asking whether the inspection itself is still necessary. The sensor data already on those vehicles answers most of what the walkthrough was designed to confirm, but because redesigning felt risky, they kept the process and wrapped new technology around it, spending money to preserve something that did not need to exist.

The organizations genuinely prepared for an autonomous future are not necessarily the ones running pilots. They are the ones where people already trust what the data is telling them, act on it and know when something needs to go up the chain. That is what this work builds.

Autonomy will reshape certain routes and use cases, and that is already happening. But the question worth sitting with right now is not whether it will arrive. It is whether your operation is genuinely built to absorb it when it does.

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?