Why bigger fleets aren’t better connected
26.8.2026 | industry matters
Why bigger fleets aren't better connectedFargo’s Commercial Director Jim Slade looks beyond the headline statistics of last month’s The UK Haulage: Technology & Resilience Benchmark Report 2026
New industry research has revealed a striking contradiction at the heart of transport technology: despite greater budgets, resources and IT capability, larger fleets are no more integrated than medium-sized ones.
The UK Haulage: Technology & Resilience Benchmark Report 2026, produced in partnership with the RHA, found that 91% of hauliers still operate below the level of seamless integration.
But perhaps the more interesting question is why scale isn’t closing the gap.
In this Industry Matters contribution, the Fargo Group’s Commercial Director Jim Slade looks beyond the headline statistic to explore where the integration challenge really lies – particularly in complex intermodal movements – and why connecting more systems may be only part of the answer.
Why scale doesn’t fix integration?
New research with the RHA found 91% of hauliers running below seamless integration. The finding sitting next to it is the one worth an hour of your time.
The number everyone will quote from the RHA research is 91%. That’s the share of road transport operators running below seamless integration, which is a large and quotable figure and has been doing the rounds since released.
However, we think the finding printed alongside might just be more useful to the industry. Integration doesn’t improve with scale. Enterprise fleets came out no better connected than medium ones.
That should stop you for a second. Bigger operators have more budget, more IT people, more governance and considerably more reason to care. If integration were a money problem, or a maturity problem, or a we-haven’t-got-round-to-it-yet problem, the large fleets would have pulled clear. They haven’t. They are sitting at the same ceiling as everyone else.
So, it is worth asking what kind of problem it actually is.
Here is a take worth considering. Integration inside a business is broadly a solvable discipline. You can scope it, buy it, resource it and finish it. The bigger challenge and the one we’d argue is fundamentally more difficult to solve is the integration between businesses. Especially in industries where the number of businesses you have to work with is significant and potentially grows faster than your capacity to connect to them.
In the middle mile, every new customer arrives with a portal. Every new subcontractor has its own way of telling you where the vehicle is, and its own idea of how quickly that update needs to happen. Every terminal runs its own booking system, its own reference format and its own view of the moment a container became available. Win more work and you add counterparties faster than you add connections. The ceiling sits at the edge of your tech stack, at the point where your data stops and someone else’s starts.
Which could explain the survey result. Scale doesn’t relieve the pressure. Scale is the pressure.
Intermodal is where this bites hardest
Anyone running an intermodal corridor will recognise this in an uncomfortable way.
A straightforward road movement might involve two companies. A single intermodal movement involves a port, a collection leg, a rail operator, at least one terminal, a delivery leg and a customer. There is a good chance no two of them are the same business, and a better chance no two of them use the same system. The handovers are not an inconvenience in the middle of the job. The handovers are the job. They are also the points where the information goes quiet.
That is why the integration ceiling is a harder ceiling in intermodal than in road. You are not connecting a fleet. You are trying to hold a common picture across organisations that have no obligation to keep you informed, no shared vocabulary, and no shared definition of success. The rail operator’s job is done when the train arrives. The depot’s job is done when the container leaves the gate. Each party is focused on their own leg, not the movement the end customer actually cares about.
The part the integration conversation keeps skipping
Suppose you fixed it anyway. Suppose every system in your business and every system belonging to everyone you work with spoke to each other perfectly.
You would have solved the plumbing. What comes out of the tap is still data. More of it, arriving faster and better organised, and none of it able to tell you which of tomorrow’s four hundred moves is the one that will hurt.
A container sitting at a terminal for six hours is either completely normal or the start of a bad week. The feed cannot tell you which. Only a record of how that terminal, that corridor and that customer usually behave can tell you which, and that record has to be long enough to know the difference between a delay and a pattern.
Tidiness of architecture matters less than the industry currently assumes. Knowing on Tuesday what Thursday looks like matters a great deal more. Connection tells you where things are. It does not tell you what to do about it, and the gap between those two things is where most of the cost in this industry lives.
There is one more result worth pulling out
Asked what they want from AI, operators named smart document processing and predictive maintenance. Those two answers map almost exactly onto the two things they said frustrate them most day to day.
This has already been described as a lack of ambition. I would call it good judgement. These are people who have been sold transformation before, more than once, and would now like something that works on Monday morning.
The same instinct is worth applying one level up. If a system tells you a vehicle is late, you have a feed, and you almost certainly already knew. If it tells you on Tuesday that Thursday’s train is at risk because the terminal has been running an hour behind all week and your customer’s booking pattern says the load will be tight, you have something you can act on. Same pragmatism. Applied to decisions rather than admin.
A question for this week
None of this makes the integration work optional. The plumbing still has to be done, and the RHA’s members are right that the effort involved is a real drag on productive time.
But it is worth being precise about the kind of integration you are buying. Connecting one more counterparty, one more time, is a bilateral answer to a multilateral problem. Do it fifty times and you have fifty things to maintain, fifty things to break, and still no single account of what happened on a movement. That is the arithmetic behind the 91%, and it is why the large fleets never pulled clear of the medium ones. They were not underfunded. They were solving it in pairs.
Getting those parties onto one record is the first half of the prize, and the smaller half. The larger half is what a shared record does over time. Fifty pairs of integrations produce fifty half-stories, none of which can tell you how this terminal behaves in the last week of the month, or which customers quietly move their booking pattern in December, or how much slack this corridor actually absorbs before a delay turns into a missed connection. A record held in common learns those things. Pairs cannot, because neither side ever sees the whole movement.
That is the difference between a system that shows you the day and a system that has an opinion about the week. One requires connection. The other requires enough history, across enough of the parties involved, to know what normal looks like and to say so early.
So before you spend another year measuring how connected you are, try this instead. Take the next fifty moves in your plan. Count how many depend on information belonging to a company that does not work for you. Then ask a harder question than whether you can see it.
Ask what your systems know today that they did not know twelve months ago. If the answer is nothing, connecting another one will not change it.




