Why freight’s middle mile still has no “Big Four”

In U.S. logistics, it’s hard to find a clearer contrast than parcel versus freight. Parcel delivery has become increasingly standardized and dominated by a small set of massive networks. Freight, the middle mile that connects suppliers, distribution centers, fulfillment centers, cross-docks, and stores, remains structurally fragmented, operationally inconsistent, and expensive to manage.

That gap is not the result of underinvestment in software alone. It’s the result of how each market is built.

Parcel consolidated because standardization made scale inevitable

The U.S. parcel market has scaled to enormous volumes, and the largest carriers move a meaningful share of that volume. Pitney Bowes reported total U.S. parcel volume of 22.37 billion shipments in 2024 (up from 21.65 billion in 2023). And in a separate report, they reported USPS handled 6.9B parcels in 2024, and Amazon Logistics handled 6.3B. That’s over half the volume.

Parcel’s operating model rewards a few things that consolidate naturally: a standardized label-and-scan unit, repeatable sortation patterns, dense route structures, and clear service definitions. When those parameters exist, the network economics and customer expectations push toward fewer, larger systems.

Freight doesn’t have that luxury.

Freight fragmentation is rooted in variance, not just competition

Freight, particularly LTL and shared flows that mix pallets, cartons, and store delivery requirements, operates under a far higher degree of variance:

--Shipment attributes vary widely (package dimensions, handling requirements, appointment constraints)

--Receiving nodes vary widely (DCs, stores, job sites, residences)

--Exceptions are common (reweighs, accessorials, missed appointments, nonstandard packaging)

--Accountability often splinters across facilities, carriers, brokers, and TMS layers

Even within a single shipper’s network, freight may be routed through multiple providers and facilities that lack visibility into each other. Freight networks often feel stitched together across vendors, contracts, and systems, with finger-pointing when things break.

This variance makes it harder for any one actor to impose standardization. And without standardization, the market can’t consolidate the way parcel does.

The carrier universe alone explains a lot

One simple indicator of fragmentation is the sheer count of operating entities. According to the American Trucking Association, citing U.S. DOT data, as of June 2025, there were almost 580,000 active U.S. motor carriers registered with FMCSA that own or lease at least one tractor.

That is the opposite of a “few key players.” It creates a long tail of capacity and a constant challenge for shippers: execution quality, visibility, and pricing discipline vary dramatically by provider, lane, and region.

The middle mile is where density, control, and economics are created

Parcel scale is created through dense sortation + linehaul. Freight scale is created through dense middle-mile transfers – where loads are consolidated, reconfigured, and moved through predictable corridors.

Attempts to unify parcel and LTL should, in theory, unlock efficiency, but they rarely work end-to-end in practice. Incumbents have still benefited from a structural advantage: pickup density and transfer density provide a buffer against underfilled trucks that smaller or newer networks cannot easily absorb.

This is one reason freight can remain “broken” even when many individual providers are competent. In freight, everyone else has to be perfect to survive, and perfection is hard when the system itself fragments accountability.

What can be done to fix freight fragmentation?

Fixing freight does not require a single monopoly network. It requires changes in how middle-mile execution is designed and governed. Three shifts matter most.

1) Treat execution as a single operating layer, not a chain of vendors

Freight networks often break down at the points of connection between facilities, carriers, and shipper systems. Fixing that requires operating models in which cross-docking execution and transportation move in lockstep. The same system, same incentives, same ownership of outcomes.

For investors evaluating logistics technology, this distinction is decisive. Software that only sits “above” the operation can improve planning and communication, but it often cannot prevent the operational blame chain. Software that is embedded into execution, creating shared truth at the handoff points, reduces fragmentation directly.

2) Build density deliberately, then expand the product surface area

Parcel networks punish thin density: if you cannot fill linehaul consistently, unit costs spike, and growth becomes price competition. The same is true in freight.

One emerging blueprint is to start with pricing and service structures that enforce discipline (for example, defined lanes and standardized units), and to diversify the sources of volume feeding those corridors (self-serve shippers, platform-embedded demand, and enterprise lanes). The strategic point is that density must compound rather than collapse.

When density compounds, additional services (including more granular pricing and mixed freight types) become rational rather than fragile.

3) Make visibility a system-of-record problem, not a dashboard problem

Freight is fragmented partly because the system of record is fragmented. When shippers cannot trust the “where is it, what is it, who touched it” trail, they spend labor reconciling disputes, chasing updates, and buffering inventory.

Technology is moving fast, but adoption is uneven. McKinsey projects that even with growth, only about a quarter of warehouses in North America will employ some level of automation by 2027. Meanwhile, in a recent podcast, McKinsey also notes that generative AI can reduce documentation lead times by up to 60% and cut logistics coordinators’ workload by 10-20% in some processes. Useful gains, but only if the underlying execution data is captured consistently.

That’s why several operators are prioritizing computer vision, scan-gap reduction, and more accurate real-time systems of record inside cross-docks, paired with workflow automation that can be introduced without rebuilding facilities.

What does this mean for the future of freight investment?

If the parcel category’s winner-take-most structure was enabled by standardization, freight’s next wave of value creation will be enabled by systems design. Investments will flow into models that unify facilities and transportation, compound density through shared corridors, and generate trusted execution data at handoffs.

The middle mile is not a footnote between the warehouse and the last mile. It is the layer where reliability, utilization, and unit economics are created, and where fragmentation is either reinforced or eliminated.

For investors, the question is less “who has the best dashboard” and more, which companies can turn freight from a chain of disconnected parties into a coherent operating system? A system where accountability is engineered into the handoff, not negotiated after the fact.