Whole companies exist just to check whether your TMS quoted the right number.
The software makes the mess. Someone else sells the cleanup. And somewhere along the way, everybody just accepted that.
Modes are the same story
There are good LTL systems that never got TL right, or never touched parcel. Not because anyone thought running three systems was good architecture — but because depth in one mode was all a focused team could realistically deliver. TL isn’t LTL with different labels. Parcel isn’t either.
For a long time I accepted that ceiling too. Here’s what changed my mind.
What changed my mind
Last week we onboarded a Canadian LTL carrier’s rating API. Their published accessorial list doesn’t match what actually prices — the real codes live in a vocabulary nobody documents. We dug it out, tested every code against live quotes, and along the way caught a bug in our own response parsing — before it ever touched a customer quote.
That was two days of work while covering other tasks. Two years ago, for us, that was a quarter — waiting for things to flush out on live loads.
The ceiling moved
I’m not claiming we’ve solved the industry. I’m saying the ceiling on what a small, focused team can cover has moved a lot — and I don’t think most software buyers have updated their expectations yet.
So here’s the question worth sitting with: if your rating reconciled to the invoice before the freight ever moved — what would you cut first?