Five scenarios for freight and logistics intermediation, the probabilities behind each, and the indicators that will tell you which future is arriving
Bottom-line first: freight intermediation is not going to disappear. The industry has a peculiar advantage in answering that question, because unlike most sectors debating what AI will do to them, logistics already ran the experiment. Convoy raised more than a billion dollars, with Bezos and Gates money behind it, on the thesis that software would automate freight brokerage out of existence. In October 2023 it collapsed. The wrong lesson to draw is that automation failed. The right lesson is that Convoy automated the wrong thing, or more precisely, automated the only thing it thought the business was, and discovered too late that matching a load to a truck was never the product.
WE PREDICT that freight intermediation in 2040 will be a genuinely better business than it is today, with higher margin per employee, stickier shipper relationships, and economics that concentrate rather than erode. It will simply be a business with far fewer firms. More than twenty thousand licensed freight brokerages operate in the United States today; the number that can credibly claim a seat in the consolidated future is a small fraction of that, and the claims are being staked in the next thirty-six months, not in 2039. Convoy’s thesis wins by 2040. It just wins inside firms that also own the functions Convoy never valued.
That conclusion comes from looking at what intermediation actually sells, function by function, and asking which functions AI absorbs and which it structurally cannot. It also comes from watching intermediation businesses that already completed this transition: travel distribution watched its commodity tier go fully algorithmic while complex itineraries consolidated around a smaller advisory layer, and media buying went programmatic on the same pattern. The sequence is consistent: the transactional middle evaporates, and the judgment layer above it gets smaller, more senior, and more profitable.
What Freight Intermediation Actually Sells
Strip the industry down, and brokerage and third-party logistics sell six things:
- capacity matching (this load, that truck, now)
- price discovery (what a lane is worth today, and what it will be worth at bid season)
- credit and payment intermediation (carriers paid in days while shippers pay in sixty)
- exception management under disruption
- capacity assurance when markets tighten, and
- blame absorption for shippers who need a named party standing behind freight worth more than the truck carrying it.
AI absorbs the first three almost completely by 2040. Matching is solved math; the pricing algorithms at the large digital freight networks already outperform mid-tier human pricing desks on commodity lanes; and payment intermediation is being rebuilt as fintech rails, with factoring, insurance, and fuel advances bundled into platforms rather than sold by people. These were the functions Convoy automated, which is exactly why Convoy’s product was excellent and its business was doomed: it competed on the three functions with no moat, in an industry where anyone can subsidize a match.
The last three resist automation for structural rather than technical reasons. Exceptions are where logistics actually lives; a dock that won’t unload, a driver who quits mid-route, a reefer unit failing outside Amarillo with $2 million of pharmaceuticals aboard. Resolving them requires judgment across incomplete information and human counterparties who are tired, angry, or unreachable. Capacity assurance in a tight market is a relationship asset that compounds over years and cannot be spun up during the crisis that makes it valuable. And when the load goes missing at 2 AM, the shipper’s VP of supply chain needs a human whose reputation is collateral. That need does not disappear when the human stops holding a rate advantage; it becomes the entire job. The 2040 intermediary is an exception-and-assurance layer, not a matching layer, and the economics of that layer favor fewer, more senior people.
The Convoy Question
Every industry has a structural question it keeps re-asking. In commercial real estate it is why no commercial MLS ever emerged. In logistics it is why Convoy failed, and the answer is worth getting precisely right, because the sloppy version of the answer (“digital freight doesn’t work”) is currently protecting thousands of brokerages that should not feel protected.
Convoy failed because it brought venture economics to the one slice of the business that was already commoditizing, during a freight recession, against incumbents whose real assets were the relationships and exception muscle Convoy’s model treated as legacy overhead. The technology worked; Flexport bought it when the company died, and versions of it now run inside the survivors. What failed was the assumption that the automatable part of the business was the valuable part.
That distinction is the entire strategic map for the next decade. The fragmentation that makes intermediation necessary is not going away; roughly ninety-five percent of trucking companies operate ten or fewer trucks, and no algorithm consolidates a physical industry that atomized. But every function that can be automated will be, and the firms treating Convoy’s collapse as permission to stand still are misreading an early arrival as a wrong address.
Five Futures, with Probabilities
Scenario 1: The Consolidated Orchestration Layer (~35%)
The base case, and the winnable one. Intermediation survives as an industry, but total headcount falls by half or more while loads per employee rise several-fold. The traditional entry tier, the track-and-trace desks and check-call floors where the industry trained its young, largely disappears into automation. A smaller senior layer operates on top of AI platforms that handle matching, pricing, tendering, and first-pass exception triage, and intervenes where judgment, relationships, and liability live.
The dividing line in this scenario is not scale, capital, or freight mix. It is whether a firm’s carrier and lane intelligence lives in institutional systems or in individual reps’ heads and personal phones. Firms in the second category dissolve when those reps retire or take their book across the street; firms in the first category absorb their volume. This scenario requires no coordination problem to be solved and no platform monopoly to emerge; it is simply every firm optimizing independently, which is why it carries the highest probability.
Scenario 2: Platform Capture (~20%)
A digital freight network or data platform wins where Convoy failed, precisely by not repeating Convoy’s mistake: instead of competing on matching, it bundles matching with the fintech layer (payments, factoring, insurance, fuel) and takes transaction economics on rails everyone else must use. Brokers become licensed operators paying rent on infrastructure they do not control. Commodity freight (dry van, standard reefer on dense lanes) trades in continuous algorithmic auction. Intermediation persists in complex freight (flatbed, hazmat, project cargo, cold-chain pharma) but the economics of the easy middle migrate to the platform even where the intermediary role survives.
Scenario 3: Principal Disintermediation (~15%)
Displacement arrives not from a platform but from the shippers themselves. The largest corporate shippers internalize AI transportation desks, run their own routing guides with algorithmic repricing, and contract directly with carriers at a scale that once required a brokerage’s network. The brokered share of contract freight falls steadily. Intermediation retreats to the mid-market shippers and irregular freight that cannot justify an in-house desk, a market that remains large but carries thinner fees. The tell for this scenario is the first Fortune 500 shipper publicly attributing a nine-figure freight-cost reduction to an in-house AI procurement desk.
Scenario 4: Full Agentic and Autonomous Displacement (~10%)
Shipper-side AI agents tender to carrier-side AI agents; autonomous trucks, already running driverless commercial lanes in Texas, carry the hub-to-hub middle mile at a cost per mile that reprices the entire market; humans are reduced to final sign-off and physical touchpoints. Technically plausible by 2040 for dense commodity lanes. But the binding constraints are stubborn: liability does not automate, cross-party data standards have defeated every previous attempt at unification, and the physical world refuses to standardize; docks jam, weather closes passes, freight arrives damaged. Meaningful probability in narrow commodity segments, near zero for exception-heavy freight.
Scenario 5: Muddle-Through (~20%)
Efficiency gains everywhere, structural change nowhere. The industry that moves everything still runs substantially on 1970s EDI standards, PDFs, and phone calls, and the fragmentation of twenty thousand brokers and hundreds of thousands of small carriers slows every transformation. Logistics in 2040 looks like logistics today with better tools and a thinner middle tier. History gives this outcome real credit; the industry absorbed load boards, TMS platforms, and an entire digital-freight funding cycle without structural displacement. But muddle-through is a probability, not a plan, and it is the only scenario in which doing nothing works.
The Land Is Being Claimed Now
Read the scenarios together and one pattern dominates. In four of the five futures, intermediation still exists and can be highly profitable; in every one of those four, the profits accrue to a far smaller set of firms than are profitable today. The commodity tier of freight migrates to algorithmic, and increasingly autonomous, rails in nearly every scenario; the only real question is how far up the complexity stack that logic climbs before it hits the trust ceiling.
The firms that survive above that line share a common architecture: carrier and shipper relationship intelligence that belongs to the institution rather than the individual, a deliberately senior exception-and-assurance layer, payment and fintech capability owned or credibly partnered, and data discipline across TMS, visibility, and pricing systems that most brokerages today cannot honestly claim. None of that is built in a quarter, which is why the land grab is now.
What to Watch
Scenario probabilities are only useful if you can tell which future is arriving. Five leading indicators are worth tracking:
- Autonomous middle-mile cost parity. When hub-to-hub autonomous cost per mile publicly undercuts team drivers on a major lane, capacity economics reprice across the whole network, not just the autonomous corridor.
- Platforms taking transaction economics. The first move by a dominant load board, visibility, or rate-data provider from subscription revenue into per-load fees is the opening act of the platform capture scenario.
- The first in-house AI transportation desk. One Fortune 500 shipper publicly crediting an internal AI procurement desk with major freight savings moves principal disintermediation from theoretical to boardroom agenda item.
- Machine-to-machine tender acceptance. The first production lane where loads are priced, tendered, and accepted agent-to-agent with no human touch will be publicized, and then quietly copied.
- Loads per employee. The industry’s quiet productivity metric. When top-quartile brokers multiply it several-fold while headcount holds or falls, the consolidation math stops being a prediction and starts being a quarterly report.
The Strategic Question
The question for a logistics leadership team in 2026 is not whether AI will change the industry; Convoy answered that, expensively, on everyone else’s behalf. The question is which scenario your firm is positioned for, and whether your current investments are building toward a seat in the consolidated future or merely making the present more efficient. Those are different projects with different architectures, and the firms that conflate them will discover the difference in the middle of a freight cycle that punishes discovery.
Innovation Vista works with logistics firms to answer exactly that question: translating what intermediation industries that already completed this transition have learned, and turning it into a positioning strategy for the consolidation ahead. If you want to pressure-test which side of 2040 your firm is building toward, that conversation is where we start.
