Your haulage company is growing, but so is the admin. How do you break the cycle?
Going from twenty to thirty vehicles usually means, in most haulage companies, an extra planner, half an extra administrative employee and an office that is still behind on Friday afternoon. That is no coincidence. This article explains why administration in transport grows in step with your fleet, and in what order you can intervene to break that pattern.
Administration grows in step because every extra trip brings the same manual steps with it. You break that in a fixed order: process, integrations, automation, custom software.
If you reverse the order and start with software, you automate the mess. Below are the steps and what each one delivers.
Administration in transport grows linearly, because every trip requires the same manual steps
Twenty vehicles that each run four trips a day make eighty trips. Every trip goes through the same steps: entering the order, scheduling, sending it to the on-board computer, printing the CMR, getting the POD back, processing hours, invoicing. Thirty vehicles make one hundred and twenty trips, and so one and a half times as much of every step.
As long as a person performs those steps, the office time required grows exactly as fast as the number of trips. That is why the rule of thumb "one administrative employee per so many vehicles" is so persistent in many companies: it holds true, as long as the work per trip stays the same. The only way to break that line is to reduce the work per trip. An integration between your TMS and on-board computer costs money once and then processes every trip, whether there are eighty or two hundred.
That distinction is the heart of the scaling question. An extra employee solves today's problem and pushes it on to the next growth step. Less work per trip solves it for every growth step after that. Our article on why transport administration takes so much time covers the seven causes one by one; here the question is the order in which you tackle them.
The problem rarely lies with the people, usually with the handovers between systems
A trip passes through at least four pairs of hands and four systems: from the planner to the driver, from the driver to the administration, from the administration to the accounting system and from the accounting system to the customer. At every handover, something is retyped, followed up by phone or checked.
Those handovers are where the time goes. The planner creates the trip in the TMS and retypes it into the on-board computer, because the two aren't integrated. The driver hands in a paper trip sheet, because the app can't capture a POD. The back office lays the charter invoice next to the trip list, because nothing does that automatically. And accounts asks on the fifth of the month whether all PODs are in, because otherwise the invoice can't go out.
A growing company suffers more from this than a small one, for a simple reason: with ten vehicles, the planner knows every driver and every customer, and catches handover errors from memory. With forty vehicles that no longer works, and handovers become more formal, slower and more error-prone. The company then adds checks, and those checks are manual work again.
The order decides whether it works: first the process, then integrations, then automation, then custom development
Companies that start automating before the process is settled end up automating the mistakes too. Companies that start with custom development before they know which steps really deviate build too much. The order below avoids both.
- The process: one order number from intake to invoice
Agree that every order gets one number that appears on the planning, the CMR, the trip sheet, the POD and the invoice. Define who enters which data and who owns addresses, time windows and rates. Set a fixed cut-off time for changes. This requires no software, but without this step no integration can know which trip belongs to which document.
- Integrations: getting your TMS, on-board computer and accounting to work together
The trip goes automatically from the TMS to the on-board computer, the status and hours come back automatically, and the completed trip goes to the accounting package as an invoice line. That removes the retyping between systems, the biggest source of duplicate work. The integrations a TMS needs are covered on our page about building a TMS; how an integration with the on-board computer works is covered on the page about a transport planning API.
- Automating recurring steps
Now that the systems know each other, you can automate the steps that recur on every trip: the CMR and weighbridge ticket are read by document recognition, the POD arrives via a driver app, and the charter invoice is automatically matched against the trips driven. Your staff review exceptions, not every document.
- Custom development where standard software doesn't fit
Only at this stage does it become clear which parts of your process an off-the-shelf package can't handle: your own customer portal, planning with sector-specific rules, or a module your TMS lacks. You then build that in a targeted way, around the integrations from step two. See building transport software and, more broadly, custom software development.
Why this order? Integrate before automating, because document recognition without an integration still needs a person to enter the result in the TMS. Automate before building custom, because most recurring steps can be solved with existing building blocks, so custom development is only needed for what remains. Reverse the order and you build at greater cost and still have manual work left over. Our page on automating logistics processes lists the building blocks we use for this.
You can tell it works from the administrative hours per trip
Hours per month rise with growth by themselves, so they say nothing about your approach. Hours per trip do: if those fall while the number of vehicles rises, the trend has been broken.
The measurement is simple. Each week, add up the hours that planners and administrative staff spend on data entry, document processing, checks and chasing up by phone, and divide that by the number of trips in that week. Do this before you start and after each step. A company with twenty vehicles that goes from forty-five minutes per trip to half an hour can, on paper, grow to thirty vehicles without extra office staff. You can work out what your current manual work costs per year with our manual administration calculator.
Does the pressure start with the planning itself, for example because two planners are working in a single Excel file? Then transport planning in Excel: when does it stop working? is the better starting point. And if you operate internationally, the eFTI Regulation from 2027 gives you one more reason to process documents digitally; read more in eFTI 2027: why invest now in digitising transport documents?.
steps, in this order: process, integrations, automation, custom software.
systems that rarely work together at a transport company: TMS, on-board computer and accounting.
an order number that follows each job from intake to invoice is the first step.
Frequently asked questions
The questions transport companies ask us most often about this.
Isn't an extra administrative employee simply cheaper than software?
In the short term, often yes. The difference is that an employee scales with the number of trips and an integration does not. At the next growth step you face the same choice again. Calculate both over several years and over the number of vehicles you expect to have, not just over the coming year.
Can we skip step two if our TMS already talks to the on-board computer?
Then part of step two is already done. Do check that the integration works in both directions (trips going out, status and hours coming back) and that the accounting system is connected as well. An integration that only sends trips leaves most of the manual work on the return side.
How do we know whether our process is "settled" enough for step two?
A practical test: can a new employee use the order number to fully trace any trip from last week, from order to invoice, without asking anyone anything? If so, the process is structured enough to integrate.
What if our TMS has no API?
Then there are three options: an export-import integration as an interim solution, an add-on alongside the TMS that takes over the integrations, or replacing the TMS. Which one fits depends on how much of your process lives in that TMS and how long you want to keep using it.
Isn't custom software only for large transport companies?
Custom software is for companies with a process that deviates from the norm, regardless of size. A company with fifteen vehicles doing refrigerated transport with temperature logging is more likely to need its own module than a company with fifty vehicles in standard distribution. Moreover, step four concerns targeted components, not a complete custom system.
How do we prevent the administration from growing again after implementation?
By keeping up the measurement from this article: administrative hours per trip, every quarter. If that number rises, a new manual step has crept in, for example a new client with its own portal or a new document type. You then tackle that step separately, rather than putting an employee on it.
Further reading
Why does transport administration take so much time?
Seven causes, from double data entry to manual checks, each with a solution.
Read more CalculatorWhat does manual administration cost per year?
Enter people, hours and wage costs to see what the manual work costs and what it frees up.
Read more PlanningTransport planning in Excel: when does it stop working?
Six signs that your fleet planning no longer scales, and four routes forward.
Read moreIs your administration growing faster than you'd like?
Briefly describe how many vehicles you have, which systems you use and where the office currently spends the most time. We'll offer our no-obligation view on the first step in the sequence above.