When an off-the-shelf TMS stops fitting a freight forwarder

Freight forwarders rarely leave a ready-made system because it is bad. They leave because the business moved and the system stayed where it was. First there is one field that does not exist. Then one sheet kept alongside the system. A year later the team works in two places and neither one is right. This is about recognising that point before it gets expensive.

What does an off-the-shelf transport management system do well?

It handles what is the same for every forwarder. Order, carrier, route, document, invoice — these steps repeat across the whole market, which makes them worth buying rather than building. A ready-made system gives you that foundation faster and cheaper than you could build it yourself.

So starting with a ready-made system is usually the right decision. The question is not whether it is good, but whether your business still fits inside what it assumes.

Which parts of the work stop fitting first?

What breaks first is not the core — it is your own processes, the ones you actually win on. In practice it is almost always the same four:

• The status structure. The system's stages are generic, while your client is asking about a specific intermediate step that does not exist in it.
• Pricing and margin. Your calculation has its own rules — surcharges, currencies, per-client exceptions — and there are no fields for them.
• Differences between suppliers. Every partner delivers data their own way, and the system expects one format.
• Client visibility. The client wants to see their own shipment, and the system has no such access.

What these four share is that none of them is about the freight. They are about the way you work — which is exactly what a ready-made system cannot know in advance.

What do the workarounds actually cost?

The cost almost never shows up on an invoice. It appears in the gaps the team fills by hand: data entered a second time into a sheet on the side, a status confirmed by message, a margin recalculated separately, and a client answered only after someone has checked three different places.

The dangerous part is not the time — it is that the truth becomes distributed. When part of the process lives in the system and part lives beside it, there is no longer one place you can trust. From that point on every decision is made from a partial picture, and mistakes surface only once they have cost something.

When we run an operational audit, most companies find 2–3 places where money was leaking that they did not know about. They are nearly always in exactly these gaps between the system and the workaround.

What does owning your own system mean?

Owning your own system does not mean rebuilding everything from zero. It means the process you compete on lives inside the system rather than beside it — with your statuses, your rules, and your suppliers' particularities.

The practical difference shows up when something needs to change. In a ready-made system you request the change and wait — or never get it, because your need is too narrow for the whole market. In your own system a change is work that can be scheduled. Our clients add 2–3 new capabilities per quarter on average — not because something was missing at the start, but because the business keeps moving.

What does this look like in a real business?

A concrete example from our own work — a car import broker whose operation is structurally very close to forwarding: many suppliers, a long shipping chain, and a client who constantly asks where their cargo is.

We connected three separate logistics providers into one dashboard and track 11 shipping stages for each provider separately — not one shared set of stages for all of them. That is precisely what a ready-made system cannot do: it has to assume one model, while in reality every partner works its own way. An import cost calculation that used to take 15 minutes per vehicle now takes seconds, and clients can see their shipment status themselves without calling.

The core platform went live on real data in about six weeks. The full story is in the AgoAuto case study, and we describe similar work for forwarding and brokerage operations on the solutions page.

When is building your own NOT worth it?

Not for everyone and not always — and that is worth saying up front. If your processes broadly match the market standard, if the team is small and the load is steady, a ready-made system will be cheaper and faster. The same applies if you cannot currently assign someone to define the process with us — without that, a custom system just becomes a more expensive copy of the same chaos.

The line is simple: your own system starts to pay off when the workarounds have become constant work rather than the occasional exception.

Where should you start?

Start not by choosing a system but by mapping your own process: which steps actually happen, where data gets entered twice, and which questions the team checks by hand. Once that is written down, it becomes obvious which part should stay in the ready-made system and which should move into your own.

We build that map together with the people doing the work, not only those managing it — the full sequence is described in how we work. If you would like to draw it together, get in touch; the first conversation is about the process, not the tools.