A $119 billion market. 155,000 employees. 2,000 terminals. One model.

The U.S. less-than-truckload market will reach $118.68 billion in 2026, growing at roughly 4–6% per year. It is the backbone of how goods move across the country — palletized freight too small for a full truck, too large for parcel.

And it is still running on infrastructure built three decades ago.

This is a data-driven look at where LTL stands today: who controls the market, where the structural inefficiencies live, what the incumbents are saying on their own earnings calls, and what a fundamentally different model looks like when you rebuild from the ground up.

1. Market Size and Growth

The North American LTL market generated $84.6 billion in revenue in 2024. The U.S. segment is projected at $118.68 billion in 2026, with a compound annual growth rate between 4.08% and 6.8% depending on the research firm.

This growth is driven by three forces:

• E-commerce fulfillment — more shipments, smaller sizes, faster delivery expectations
• Nearshoring — manufacturing moving closer to consumption, generating more regional freight
• Inventory strategy shifts — companies holding less inventory, shipping more frequently in smaller quantities

The market is large. The market is growing. And the carriers serving it have barely changed how they operate.

2. The Incumbents: Six Public Carriers, $28 Billion in Revenue, One Model

The top 25 LTL carriers control 91% of a $52.8 billion addressable market. Here is what the top 6 publicly traded LTL carriers look like:

Carrier	2024 LTL Revenue	Terminals	Employees	Shipments/Day	Operating Ratio
FedEx Freight	$9.4B	355	~39,000	~94,000	~80%
Old Dominion	$5.8B	~260	22,522	47,288	73.4%
XPO	$4.9B	614 locations	~38,000	51,508	85.0%
Saia	$3.2B	214	15,000+	35,390	85.0%
TForce Freight	$3.1B	~658	27,205	~25,000	88.3%
ABF (ArcBest)	~$2.4B	~240	~14,000	declining	91.2%

Six carriers. Over $28 billion in combined revenue. 155,000+ employees. Nearly 2,000 terminal facilities. Tens of thousands of owned tractors and trailers.

Every single one operates the same hub-and-spoke terminal model. Every single one is capital-intensive, labor-heavy, and structurally dependent on terminal utilization to make money.

When Yellow collapsed in 2023, its $5 billion in annual revenue got absorbed by these players. The result: higher concentration, less competition, and a pricing structure that favors the incumbent.

3. What the Incumbents Are Saying on Their Own Earnings Calls

The most revealing data about the LTL market does not come from industry reports. It comes from what carrier executives tell their own investors. Here is what they said in their 2024 earnings calls:

Old Dominion: Volume Down, Costs Up, Waiting for Recovery

Old Dominion is the best-run LTL carrier in the industry — a 73.4% operating ratio that no one else comes close to. And even they are struggling with the structural math.

• Q4 2024 tonnage per day decreased 8.2%. Revenue fell 7.3% YoY.
• Operating ratio deteriorated 410 basis points in Q4, from 71.9% to 75.9%.
• CEO Marty Freeman: The revenue decline caused "a deleveraging effect on many of their operating expenses." Overhead costs rose 140 bps. Depreciation rose 70 bps. Insurance and claims rose 100 bps — all as a percentage of revenue.
• $771 million in capital expenditures in 2024: $350M on terminal real estate, $325M on tractors and trailers, $75M on IT.
• Core cost-per-shipment inflation projected at 5–5.5% even in a flat volume environment — driven by employee benefits, insurance, equipment, and real estate.
• On the earnings call, management admitted: "We certainly feel like the stars are coming into alignment, but we felt that way before."

The bear case on ODFL is simple: when you spend $771M/year maintaining a terminal network and volumes decline, fixed costs crush your operating ratio. The best LTL carrier in history went from 71.9% OR to 75.9% in two quarters — not because they operated poorly, but because the model requires volume to function.

XPO: $7 Billion in Debt, 28 New Terminals, Still 1,000 Basis Points Behind ODFL

XPO has made the most aggressive push to modernize traditional LTL through technology. Their AI platform reduces empty miles by 12% and drives demand forecasting 90 days out. But the underlying model is the same.

• CEO Mario Harik: "We're currently in a historically soft freight environment."
• January 2025 tonnage was down 8.5% from the prior year.
• Debt-to-equity ratio: ~2.1 (Q4 2024 balance sheet). Interest expense projected at $205–215 million for 2026.
• Capex surged from 3.8% to 14.6% of revenue between 2018 and 2024 — nearly quadrupling the capital intensity.
• 28 new service centers being ramped, each burning cash before reaching maturity.
• Operating ratio of 85.0% — roughly 1,000 basis points worse than ODFL despite heavy tech investment.

The XPO story reveals the limits of applying AI to broken infrastructure. You can optimize linehaul routing and reduce empty miles, but you still operate 614 locations, employ 38,000 people, and carry ~$3.3 billion in long-term debt. The AI makes the terminal model slightly better. It does not replace the terminal model.

Saia: Growing Fast, Bleeding Cash

Saia opened 21 new terminals in 12 months — unprecedented in the company's 100-year history. The cost has been enormous.

• New terminals operate at a 95% operating ratio vs. 82.2% for mature locations — a 1,300 basis point gap.
• Cash dropped from $296.2 million to $19.5 million in one year — a 93% decline.
• Capital expenditure: $550 million on real estate plus $400–450 million on equipment.
• Salaries, wages, and benefits increased 8.7%. Claims and insurance up 16.6%.
• Q1 2024 OR hit 91.1% — the worst since the pandemic.
• Morgan Stanley downgraded to Underweight, slashing the price target and noting earnings estimates were "nearly halved over the past year."

CFO Matt Batteh defended the expansion: "There's cost associated with them, but there's also a cost of not being in the market." The question is whether spending $1 billion/year on terminal expansion in a freight recession is the right infrastructure bet — or whether the money is going into the wrong type of facility entirely.

TForce Freight: "A Disaster"

TFI International acquired UPS Freight in 2021. Three years later, CEO Alain Bedard called Q4 2024 "a disaster."

• US LTL operating ratio hit 97.3% in Q4 — a 630 basis point deterioration from the prior year. Q1 2025 deteriorated further to 98.9% — essentially breakeven.
• Bedard on density: "Our density is [expletive]." And: "Density is the name of the game."
• Bedard on the business: "TForce is a big rock in my shoe, no doubt about it."
• Claims ratio: 0.9% of revenue — Bedard called it "unacceptable," noting: "We just throw money out the door."
• Billing systems: "It's unimaginable that in 2024 we still have issues billing customers, but it's a fact."
• 35% excess capacity across the network, combined with low density — a structural mismatch.
• Customer churn: "We lost so many of the small and medium-sized accounts, and we replaced them with corporate accounts with sometimes negative margins."

TForce is the clearest example of what happens when you inherit legacy LTL infrastructure and try to fix it from the inside. Three years and billions in investment later, the CEO describes the business in unprintable terms.

FedEx Freight: Being Spun Off Because the Model Doesn't Work Inside FedEx

FedEx announced in December 2024 that it would spin off FedEx Freight as a standalone public company — an implicit admission that LTL does not fit inside a parcel-first organization.

• Daily shipments declined ~6%. Revenue declined 5%.
• Adjusted operating profit dropped from $261 million to $134 million — nearly halved.
• CFO John Dietrich: "A major pressure is coming from a weaker market consistent with LTL industry challenges."
• $152 million in one-time spin-off costs.
• The spin-off rationale: FedEx Freight was "not fully appreciated" within FedEx, trading at ~13x forward estimates vs. 20x+ for LTL peers.

The FedEx Freight spin-off is the market telling you that traditional LTL infrastructure is worth less than the sum of its parts when bolted onto a parcel network. The question is whether it is worth more as a standalone — or whether the model itself is the problem.

ArcBest/ABF: Declining Tonnage, Rising Costs, CEO Retiring

• Full-year tonnage shipped down 14.3%. Q2 tonnage per day dropped 20.3%.
• Operating ratio: 91.2% — meaning $0.91 of every dollar goes to costs.
• Operating income plunged 40% in Q4.
• Insurance costs spiked $9 million, adding 160 basis points to the OR.
• Long-term OR target: 87–90% by 2028 — acknowledging they are currently well above that.
• CEO Judy McReynolds announced retirement at end of 2025, creating leadership transition risk.

ABF's tonnage decline is partially a freight recession story, but it is also a structural story: the company lost Yellow freight it had gained in 2023, its shipment weight per load is declining, and the union contract rate increases continue regardless of volume.

4. The Structural Problem: Terminals, Touches, Dwell

Here is how every one of those six carriers moves a typical LTL shipment:

1. Pickup — freight loaded onto a local P&D truck
2. Origin terminal — unloaded, sorted, consolidated with other freight
3. Line-haul — loaded onto a long-haul trailer, moved to destination region
4. Destination terminal — unloaded, sorted again, staged for delivery
5. Delivery — loaded onto a local truck, delivered to consignee

That is a minimum of 5 freight touches — 5 times human hands or forklifts move your pallets. Industry estimates put the average at approximately 4–5 handling events per shipment. Every touch is a chance for damage, delay, or loss.

At a traditional LTL terminal, freight sits for 2 to 5 days depending on volume, destination, and network efficiency. No public carrier discloses average terminal dwell time — it is tracked in the confidential NMFTA benchmarking program but never published. The fact that no carrier reports this number tells you something about what the number is.

The capital required to maintain this infrastructure is staggering:

Carrier	Annual Capex	What It Buys
Old Dominion	$771M	Terminal real estate, tractors, trailers, IT
XPO	8–14% of revenue (~$500–700M)	New service centers, fleet, technology
Saia	~$1B	21 new terminals, equipment
FedEx Freight	Not disclosed separately	355 terminals, 39,000 employees
TForce	Not disclosed separately	Maintaining 658 facilities
ArcBest	$288M	Equipment, real estate, Yellow facility acquisitions

Combined, the top 6 LTL carriers spend $2.5–3 billion per year in capital expenditure just to maintain and marginally expand their terminal networks. This money goes into buildings, trucks, and trailers that depreciate every year and require replacement on a 7–15 year cycle.

5. What the Data Says About Damage and Delays

Damage

The industry-wide LTL damage rate is 1.24% — roughly 1 in every 80 shipments results in a damage or loss claim. The average cost per claim is approximately $1,796.

Among the public carriers, only two disclose their claims ratios:

Carrier	Claims Ratio	Source
Old Dominion	Below 0.1%	Q4 2024 earnings
XPO	0.2% (down 80% in 2 years)	Investor presentation
TForce	0.9% of revenue	CEO Bedard, Q4 2024 call — called it "unacceptable"
Saia	Not disclosed	—
FedEx Freight	Not disclosed	—
ArcBest	Not disclosed	—

The carriers that do not disclose their claims ratio are not hiding good news. TForce's 0.9% — which the CEO publicly called "unacceptable" — is likely better than what some non-disclosing carriers run.

The root cause is structural: more touches, more time in terminals, more opportunities for damage. No amount of claims management fixes a 5-touch, multi-day dwell process.

On-Time Performance

The industry average for on-time delivery in LTL is 82%. Old Dominion reports 99%. No other public carrier discloses their OTD rate.

For shippers serving big-box retailers with 95%+ on-time-in-full requirements, this is a fundamental problem. You cannot meet a 95% OTIF standard with an 82% carrier.

6. The Cross-Dock Alternative

A cross-dock is not a terminal. A terminal stores, sorts, and stages freight over multiple days. A cross-dock receives freight in one door, sorts it, and sends it out another door — ideally within hours.

	Traditional Terminal (ODFL, XPO, Saia)	Cross-Dock Model
Freight touches	5+ (industry est: 4–5)	2
Dwell time	2–5 days (not disclosed by any carrier)	Under 1 day (best facilities)
Facility cost	$771M+/year capex across top carriers	Fraction of terminal cost — no storage infrastructure
Employees	14,000–39,000 per carrier	Zero owned fleet, zero employed drivers
Storage	Yes — freight staged for days	No — flow-through only
Damage risk	Higher (more handling)	Lower (fewer touches)

What 58,000 shipments show

Across 58,224 shipments processed through cross-dock facilities in a network built on this alternative model:

• Best facility dwell: 0.67 days (Chicago) — 85.5% of freight processed same-day
• Second-best: 0.89 days (Columbus) — 80.4% same-day
• Los Angeles facility dwell dropped from 1.76 days to 1.32 days over 9 months — a 25% improvement while volume increased

The damage rate across 641,841 completed shipments in this model: 0.81% — 35% lower than the 1.24% industry average.

This is not incremental improvement layered onto a terminal. It is a structural reduction in damage driven by fewer touches and less dwell, in a facility that costs a fraction of a traditional terminal.

7. What AI Changes (And What It Doesn't)

70% of transportation and logistics companies now report adopting AI. XPO's AI platform reduces empty miles by 12%. Old Dominion is investing $75 million per year in IT. Saia is deploying dynamic pricing. Everyone is doing something.

But here is the critical distinction: every one of these AI investments is being applied to the same terminal infrastructure.

XPO uses AI to optimize linehaul routing across 614 locations. The terminals still exist. The freight still dwells. The 5 touches still happen. The AI makes the existing model 5–10% more efficient — it does not change the model.

Where AI creates structural change in LTL:

• Pricing at density — processing millions of quotes to find where density exists and pricing lanes accordingly. One network has processed over 11 million freight quotes, building pricing models that improve as lane density increases. That is not a tariff with a discount. It is a system that gets cheaper as volume grows.
• Automated quality monitoring — flagging late pickups, missed scans, route deviations, dwell anomalies, and temperature exceptions in real time. Compare this to TForce, where the CEO described billing systems as "unimaginable" in 2024.
• Carrier vetting at scale — continuously evaluating 22,000+ carriers on safety scores, insurance, on-time performance, and damage history. Compare this to Saia managing 15,000 employees or XPO managing 38,000.
• Cross-dock orchestration — matching inbound freight to outbound capacity in real time. This is how a facility achieves 0.67-day dwell instead of 2–5 days.

The question is not whether AI will transform freight. It is whether AI gets applied to a $771M/year terminal network that is structurally unchanged from the 1990s — or to infrastructure that was built for AI from the beginning.

8. The Data

9. The Structural Comparison

Dimension	Legacy LTL (ODFL, XPO, Saia, etc.)	Cross-Dock + AI Model
Annual capex	$300M–$1B per carrier	Fraction — no owned fleet, no terminals
Employees	14,000–39,000 per carrier	Zero owned drivers, zero terminal workforce
Owned tractors	6,600–14,500 per carrier	Zero — 22,246 carrier partners
Terminal count	214–658 per carrier	50+ cross-dock facilities
Facility type	Storage + consolidation terminals	Flow-through cross-docks
Freight touches	4.2–5+	2
Dwell time	2–5 days (undisclosed)	0.67–1.4 days (disclosed)
Damage rate	1.24% avg; 0.9% TForce (called "unacceptable")	0.81% across 641K shipments
Volume trend	ODFL: -8.2%, ABF: -14.3%, FedEx: -6%	7x growth in 21 months
Cost inflation	5–8.7% annually (labor, insurance, equipment)	Costs decline with density
Revenue model	Tariff + discount negotiations	AI-driven density pricing (11M+ quotes)
Equipment	Trailer-only (53ft)	Multi-modal: box trucks, cargo vans, trailers, reefer

9. Where the Market Goes

The LTL market is bifurcating.

Legacy carriers are locked into their infrastructure. ODFL will continue running the best terminal network in the industry — and will continue spending $771M/year to maintain it. XPO will continue applying AI to 614 locations and carrying ~$3.3B in long-term debt. Saia will continue opening terminals at 95% OR and burning cash. TForce will continue describing their own business in unprintable language. FedEx Freight will become a standalone company and try to prove the terminal model works outside of FedEx.

None of them can switch to cross-docks. The capital cost of their terminal networks is on their balance sheets. The employees are under contract. The operating model requires terminal utilization to generate returns. The switching cost is not financial — it is existential.

New entrants are building cross-dock networks from scratch — acquiring facilities, deploying carrier networks managed through driver apps and AI, and competing on structural cost advantages rather than incremental software improvements.

The shippers who move first onto dense lanes in these new networks lock in cost advantages that compound over time. As more freight flows through a lane, the cost per shipment drops. Late movers subsidize early movers.

This is not a technology trend. It is an infrastructure replacement cycle. The incumbents see it on their own earnings calls — they just cannot do anything about it.