What a Freight Broker's Manual Load Matching Costs Per Rep, Per Day

The number you were quoted is in the wrong unit

You have probably seen the math already. A brokerage moving 100 loads a week is told that moving off manual dispatch will save 65 to 100 dispatcher-hours a month. At a fully loaded rate of 32 dollars an hour that comes to 2,000 to 3,200 dollars a month. The same guide puts the wall for spreadsheet-and-phone operations at around 40 to 50 loads a week, which is about where a lot of independent brokerages sit when they first go looking for software.

Those figures are from a TMS comparison page, and we would treat them as vendor arithmetic rather than a benchmark. But the deeper problem is not whether the numbers are right. It is that hours are the wrong unit, and the people producing the number chose it because their software is priced per seat. If you sell seats, you cost the problem in seat-hours. That is not dishonest. It is just not your problem.

Your problem is coverage, and coverage is not a function of hours. It is a function of which loads a rep chooses to spend those hours on.

Reps do not fail at random

Watch a carrier rep work a board by phone for a day and a pattern shows up within the first hour. They do not work the loads in the order they arrived, and they do not work them in the order of margin. They work the ones they know how to cover.

That is rational. A rep is measured on loads covered and on not getting screamed at, so the first hour goes to the Chicago to Columbus van load with three carriers who always take it. The second hour goes to the next easiest thing. The odd lane out of a rural origin, the reefer with a four-hour delivery window, the shipper who called at four in the afternoon for a next-morning pickup: those move to the bottom of the pile, and on a busy day the bottom of the pile is where they stay.

So the cost of manual matching does not land evenly across the book. It lands entirely on the hard freight. The easy loads would have been covered under any system. The hard ones are the ones that fall off the board at six, get covered late at a panic rate, or get handed back to the shipper.

Here is the part that should bother you. Those are also the loads with the margin in them. FreightWaves has documented a strong inverse relationship between tender lead time and transactional broker margin: the shorter the shipper's booking window, the less price sensitive they are, because a load tendered days out is usually sitting under a contract commitment and nobody is paying a spot premium on it. Loads tendered same-day or next-day carried as much as a 27 percent premium over loads tendered with more than four days of lead time. Specialised and expedited freight runs a similar story, with brokers and trainers describing 18 to 25 percent margins against a 12 to 15 percent norm on standard van.

Put the two facts together and the shape of the loss is clear. Manual matching spends its capacity on the freight that needs it least and rations it on the freight that pays the most. The dispatcher-hours number cannot see this, because a rep who covers fifteen easy loads and drops the two hard ones has, by the hour, had a productive day.

What a rep-day is actually worth

We would cost a rep-day differently, and it takes a spreadsheet you already have rather than a demo.

Take a month of loads. Sort them into quartiles by how hard they were to cover. If you have no field for that, use proxies: lead time at tender, lane frequency in your own history, equipment type, and whether the pickup window was under four hours. The bottom quartile on those measures is your hard freight.

Now, for that quartile only, ask three things. What share was covered at all. What share was covered inside the first two hours of being posted. And what the realised margin was on the ones you got, versus the ones that went late.

That is your rep-day, costed in the unit that matters: hard loads covered per hour of rep time. Most brokerages we talk to have never cut the data this way, and the reason is the same one that makes it worth doing. The board shows loads, the rep shows loads covered, and the comp plan counts loads. Nothing in the operation is measuring whether the difficult loads are getting worked.

Do not expect the number to be flattering. And notice what it tells you, because it decides which purchase you should make, if any.

If the bottom quartile never gets worked, software is not the fix

There are two very different results from that exercise, and they point at different remedies.

The first is that the hard freight does get worked, just slowly and inconsistently, because reps are buried in the routine loads that could have been handled by a rule. That is a genuine automation case. Anything that takes the repeat lanes with known carriers off the phone and into a workflow will release rep hours, and in this version those hours flow to the hard freight because the reps already wanted to work it. The vendor's dispatcher-hours number is, in this scenario, roughly the right idea with the wrong price on it.

The second result is that the bottom quartile is not being worked at all. Not late, not badly. Not at all. The loads sit until they age off or the shipper pulls them. Reps step over them because working them is unrewarded, because nobody on the desk owns the odd equipment or the rural lane, or because the comp plan makes a fifteenth easy cover worth more than a first hard one.

Automation does not fix the second problem. It makes it worse. Free up ten hours a week on a desk that has been trained to avoid the hard freight, and the desk uses those ten hours to cover more easy freight, faster. The load board looks healthier and the margin mix does not move. You have paid for throughput on the loads that were never the constraint. We wrote about this general failure mode in when AI is the wrong answer to an operations problem, and freight matching is one of its cleanest examples: the tool works, and it works on the wrong loads.

What the second result calls for is a desk structure change. Someone owns the hard freight, is measured on it, and is paid for it. Whether that is a dedicated desk, a rotation, or a comp adjustment is a question we answer differently at every brokerage, because it depends on how much hard freight you actually have and how much of your margin it carries. But it is a people and incentives decision, and no seat licence contains it.

The document pile is real, and it is a separate problem

The vendor case for load matching software often folds in a second cost, and it is worth separating. At 20 loads a day a brokerage is handling something like 80 to 100 documents a day: rate confirmations, BOLs, PODs, carrier packets, most of them arriving by email in whatever format the carrier prefers. That is a real burden and it scales worse than matching does, because every load generates paperwork whether it was easy or hard to cover.

Document handling is a good automation candidate on its own terms. It is high-volume, low-judgment, and its cost does not concentrate on any particular freight. But it is not the same project as load matching, and bundling the two into one dispatcher-hours figure hides the fact that only one of them has a coverage problem attached. If your reps are drowning in paperwork, fix that. It will not, by itself, get a single hard load covered sooner, and the business case should not claim that it will. Our note on how vendors calculate automation ROI covers the general habit of putting two unrelated savings in one number.

Why this is getting more expensive, not less

The market is currently pushing the cost of unworked hard freight up. In the Bloomberg Intelligence and Truckstop survey of 141 brokers published in August, 86 percent said finding capacity has become tougher, and 72 percent expect it to tighten further over the next three to six months. At the same time, 43 percent reported lower gross margin in the first half of 2026 than in the back half of 2025.

Read those together. Capacity is harder to find, so the hard loads are getting harder. Margins are compressing on the routine freight, so the premium loads matter more to the book. A desk that quietly declines its bottom quartile is declining a growing share of its available margin, in a market where the easy loads are paying less than they did a year ago.

That is the real cost of manual matching. Not the hours. The freight the hours never reach.

What we can and cannot tell you from here

What we can say in writing is where the cost sits and how to measure it. What we cannot say is what your bottom quartile looks like, how much of your margin it carries, or whether your desk is slow on hard freight or simply not working it. Those answers live in your TMS history and in a few conversations with the people on the phones, and they decide whether you need software, a different desk, or both.

If you have been quoted a dispatcher-hours saving and something about it feels like it is measuring the wrong thing, talk to us. We will cost your rep-day on the freight that actually pays, and tell you plainly whether the tool you are looking at would ever touch it.