Key Points
- Most enterprise shippers are somewhere in the middle of the data maturity curve — not failing, but not competitive either. In this freight market, the middle is a losing position.
- The hard questions, why is cost per unit weight up, how much of it is fuel, what's the breakdown by lane and carrier, require charge-level data and need to be answered weekly, not monthly.
- AI is a real accelerant for this kind of analysis, but only once the data foundation is in place. AI running on fragmented, unnormalized data produces wrong answers faster.
- The shippers managing this market best made the decision to operationalize their freight data before conditions got this difficult. The window to do the same is narrowing.
When things are going well, companies get comfortable. Data reviews happen monthly. Budget variance gets reviewed quarterly. The reports land in someone's inbox, get skimmed, and move to the archive. Nobody gets fired for that in a stable market.
This is not a stable market.
Diesel is up 45% from pre-conflict levels. The LMI Transportation Capacity index hit 28.4 in April, the second-lowest reading in the index's history. Aggregate logistics costs are at 242.4, the highest since the peak of post-pandemic inflation in 2022. And as of late May, nothing has changed the trajectory.
In a market moving this fast, a data strategy built for stability is not just suboptimal. It's actively costing you money.
Where Most Companies Actually Are
I've been spending a lot of time with customers over the past several weeks, and I want to be honest about what I'm seeing. Most enterprise shippers are not in bad shape on data in an absolute sense. They have systems. They have reporting. They have people who understand the numbers. But most of them are also not on the right side of the maturity curve, and in conditions like these, that gap becomes expensive very quickly.
The maturity curve isn't complicated to understand. On the left side, you're maintaining. Data exists somewhere, in multiple systems, with inconsistent formats across regions and carriers, and getting a meaningful answer to a meaningful question takes weeks. In the middle, you're optimizing. Data is more consolidated, reporting is more accessible, and you can answer most operational questions if you're willing to wait a few days and do some manual work. On the right side, you're leading. The data is normalized, centralized, and accessible in near real time, and the questions that used to take weeks to answer now take hours.
The right side is where AI becomes genuinely useful. And that's not a coincidence.
Why AI Doesn't Fix a Data Problem
There's a version of the AI conversation happening in the supply chain right now that I find frustrating. It treats AI as the solution to a visibility problem, as if deploying the right model will surface insights that your existing data infrastructure can't produce. That's not how it works.
AI is an accelerant. It takes analysis that used to take a team of analysts several days and compresses it into hours. It allows you to run scenarios more frequently, spot anomalies faster, and get to a decision with less lag between the question and the answer. That is genuinely valuable, especially in a market where conditions are changing week to week.
But AI running on fragmented, unnormalized data produces faster wrong answers. If your transportation spend data sits in silos across five regional systems, three carrier EDI formats, and two legacy ERPs that were never fully integrated, putting an AI layer on top of that doesn't fix the fragmentation. It just automates the confusion.
The prerequisite is always the same: consolidated, normalized, charge-level data in a single place. That's step one. Understanding what it's telling you is step two. AI is step three, the accelerant that lets you run step two at a frequency and depth that wasn't previously possible.
The shippers who are navigating this freight market with the most clarity skipped ahead to AI because they had already completed step one. Everyone else is still working on the foundation.
The Question You Should Be Able to Answer in Under an Hour
Here's the test I use in customer conversations. It's a single question: why is your cost per unit weight up this month, and how much of it is fuel?
That question sounds simple. It isn't. A real answer requires breaking down your total transportation spend by mode, then by lane, then by carrier, then by charge type, and isolating the fuel surcharge as a percentage of each. Then you need to compare that against what you were paying sixty days ago before the diesel spike, and against what your contracts actually stipulate.
If you can answer that question accurately in under an hour, you are on the right side of the maturity curve. You have what you need to make good decisions fast, know which levers to pull, go into carrier negotiations with data behind you, and update your cost assumptions as the market moves.
If it takes your team a week, or if the answer requires stitching together reports from multiple systems that were never designed to talk to each other, that's the gap. Not the freight market, not your carrier relationships, not your modal mix. The gap is the time between the question and the answer.
In a market where diesel prices moved 52% in six weeks, a week-long data pull is not fast enough.
What "Operationalizing Data" Means
I use the phrase "operationalizing data" a lot, and I want to make sure it lands concretely rather than as a consulting abstraction.
It means freight data isn't something you consult when building a budget or preparing for a carrier negotiation. It's fundamental to how you run the business week to week. The metrics are visible, up to date, and owned by someone with the authority to act on them. The analysis isn't a project. It's a standing process.
It also means the use cases are specific. Network optimization is a use case. Contract optimization is a use case. Carrier scorecard management is a use case. Each one has a functional owner, a data requirement, a decision frequency, and a measurable outcome. You prioritize them by return on effort, because not every use case has the same payoff, and not every company has unlimited capacity to pursue them simultaneously.
Contract optimization is the one I keep coming back to as the highest near-term return for most shippers. It requires good data, solid negotiating strategy, and the discipline to run the math before you walk into the room. But the capital outlay is low compared to something like a network redesign, and the savings opportunity is real, especially in a market where fuel surcharge structures and accessorial frameworks are being actively repriced by carriers who know they have leverage.
If I had to pick one use case to master right now, that's the one.
The Window Is Getting Shorter
There's a timing reality here that I don't want to understate. The shippers best positioned in this market made data investment decisions months or years ago, before the current disruption, in what felt, at the time, like a stable environment where the urgency wasn't obvious. They did it because they understood that freight data infrastructure is not something you build in response to a crisis. It takes time, and the value compounds as the data accumulates and the organization learns to use it.
The shippers who are struggling right now are the ones who deferred that investment because the market was soft, costs were manageable, and the pressure wasn't acute enough to justify the effort.
The pressure is acute now. The market has made the case more clearly than I ever could in a customer meeting. And the window to get ahead of this, to build the foundation before the next cycle of disruption, is shorter than it was six months ago.
The freight market will eventually normalize. Fuel will come down. Capacity will loosen. The LMI numbers will soften from their current extremes. When that happens, the shippers who used this period to get their data infrastructure right will be set up to capture the upside of the next cycle faster than everyone else.
The ones who waited will be starting the same conversation over again, just with different numbers on the page.
Steve Beda is Executive Vice President of Customer Advisory at Trax Technologies. He oversees customer programs and advises enterprise shippers on navigating disruption across transportation and logistics markets. This article draws on data from the April 2026 Logistics Managers' Index (Colorado State University / CSCMP), the U.S. Energy Information Administration, and Trax's monthly Freight Market Report.