Your Dispatchers Are Guessing Routes for 250 Loads a Day. Here’s What That’s Actually Costing You.

Right now, somewhere in your dispatch office, someone is staring at a whiteboard or a spreadsheet trying to figure out which driver should take which load. They’re making judgment calls based on habit, memory, and gut feel, not math. And on a busy day, they’re doing this 250 times.

They’re not bad at their job. They’re doing something that no human brain is built to do well: optimizing dozens of variables across hundreds of loads in real time, while the order book keeps changing underneath them.

This isn’t a people problem. It’s a systems problem. And for fleet operators running private trucks alongside brokered loads across multiple regions, it’s one of the most expensive invisible costs in the business.

The Real Cost of Manual Routing

Let’s talk about what’s actually happening on the ground. A typical mid size fleet operator in this space runs two sides of the business: a bulk private fleet operation with 60-plus trucks doing waste removal and raw material procurement from mills, and a packaging division running brokered rates across multiple provinces and states.

On any given day, dispatchers are managing three categories of work: inbound pickups from mills, transfers between production sites, and outbound deliveries. Some loads have rigid time windows between 9 and 11 this morning. Others are flexible five loads by the end of week, don’t care when. And then there’s the intraday on demand mill service, which makes up 30 to 40 percent of the workload and can shift hour by hour.

The dispatchers know their routes. They know their drivers. But when Tyler’s planning 250 deliveries and a mill calls at 10am saying a bin is filling up fast and they need another truck now, he’s making shortcut decisions. He’s sending Billy to the same location Billy always services, even if another driver is five minutes away and heading in that direction already. The load gets done. But the truck runs empty on the way back. Multiply that by dozens of loads a day, and you’re bleeding margin on every route.

"On a busy day, dispatchers wind up making shortcut decisions. They send the driver who ‘always does that route’ instead of the driver who’s five minutes away heading in the right direction. Every one of those shortcuts is margin you’re leaving on the road."

The Empty Leg Problem Nobody Quantifies

Here’s the math that most fleet operators never run. Your truck delivers at point B, then picks up at point C, then runs back. But the return leg is empty. And the flexible delivery that could have filled it the one that’s due Friday but the truck is five minutes away on Wednesday nobody thought to slot it in. Because nobody could see the full picture.

When you’re running $10 million in annual trucking costs and your dispatchers are making heuristic based decisions instead of optimized ones, the empty miles add up fast. Industry data consistently shows that route optimization can reduce fleet costs by 10 to 20 percent. On a $10 million operation, that’s $1 to $2 million in annual savings sitting on the table.

The opportunity isn’t just about cutting costs on existing routes. It’s about the backhaul pairings your dispatchers never see. Your truck is running empty from a delivery but there’s a flexible pickup nearby that isn’t due until Friday. A human dispatcher doesn’t make that connection because they’re already three calls deep into the next crisis. An optimization algorithm makes it in milliseconds, because it’s looking at the entire order book at once.

Why Off the Shelf Platforms Miss the Point

If you’ve evaluated Trimble, Descartes, or any of the big TMS platforms, you already know the pitch: daily route optimization, driver tracking, accounting reconciliation. And for a lot of standard trucking operations, that works fine.

But for a fleet with complex, mixed operations, the standard platforms hit a wall. Here’s where they break down:

• Daily optimization isn’t enough. A platform like Descartes can optimize tomorrow’s routes beautifully. But it can’t look at your full weekly order book and figure out that the flexible Friday delivery should actually get slotted into Wednesday’s route because the truck is already in the area. And when three mill pickups shift from afternoon to morning with an hour’s notice, there’s no intraday re-optimization. You’re back to manual gymnastics.
• Complex business rules don’t fit the template. After delivering product A, you can’t deliver product B without washing the trailer. Certain trailer types can only go to a select subset of customers. Cross-contamination rules, trailer configuration constraints, product compatibility these aren’t edge cases in your business. They’re the business. And adding them to a rigid platform is either impossible or ruinously expensive.
• You’re paying for features you don’t need. The bigger platforms bundle accounting reconciliation, compliance modules, and functionality you’ll never touch. At $150,000 for implementation plus $250,000 a year in licensing, you’re subsidizing someone else’s product roadmap.
• Vendor lock in kills future flexibility. When you acquire a new business that doesn’t route the same way and for a company doubling every two to three years, that’s not hypothetical you need a system that adapts. Multi year contracts with rigid platforms don’t give you that.

Stat: Fleet operators who implemented AI-driven route optimization report 10–20% reductions in trucking costs, with the highest gains coming from backhaul matching and flexible time-window utilization. (McKinsey)

What AI-Powered Routing Actually Looks Like

This is where most people picture a magic button that solves everything overnight. It’s not that. It’s plumbing good plumbing that connects the systems you already have and makes them work together intelligently.

Here’s what the actual workflow looks like when you build it right:

• Orders flow in from your ERP Microsoft Business Central, in most cases with all the context attached: delivery windows, product types, trailer requirements, customer constraints. No re-keying. No copy-paste.
• The routing algorithm takes the full order book not just tomorrow’s loads, but the entire forward looking schedule and optimizes across every variable: driver location, hours of service, trailer configuration, cross contamination rules, flexible versus rigid time windows, even ferry schedules if you’re operating coastal routes.
• When a mill calls at 10am and says they need another truck, the system re-optimizes in real time. It doesn’t just find the nearest driver. It recalculates the entire routing picture to figure out the least disruptive way to handle the new load while keeping everything else on track.
• Drivers get their assignments through a mobile app that actually works not a cumbersome platform where loads get stuck and can’t be reassigned. They see their full day. They accept and complete loads. And the completed shipments push back into the ERP automatically for cost assignment.
• Costs flow back onto the right sales orders, transfer orders, and procurement orders. Tyler stops guessing at the margin. He sees real numbers.

The key difference between this and a generic TMS is that every variable that matters to your business is captured up front and baked into the optimization. Trailer types. Wash requirements. Driver familiarity. Product compatibility. Flexible delivery windows stacked against rigid deadlines. If the algorithm doesn’t have those variables from day one, you can’t add them later the foundation has to be right.

What This Means for Your Dispatch Team

Let’s be clear: this doesn’t replace Tyler and the dispatch team. It gives them superpowers.

Instead of mentally juggling 250 loads and making gut feel decisions under time pressure, Tyler is reviewing optimized route plans and making override decisions where human judgment actually matters. The algorithm handles the math. The dispatcher handles the exceptions.

The downstream effects are significant:

• Empty legs drop because the system pairs backhauls with flexible deliveries that your dispatchers never would have connected manually.
• Mill service response time improves because re-optimization happens in minutes, not hours of phone calls and whiteboard shuffling.
• Drivers stop getting frustrated with cumbersome apps that only show them one load at a time. They see their full day and can plan accordingly.
• Sales gets visibility into what dispatch is actually doing, and dispatch gets visibility into incoming orders. The communication gap between those teams which is a constant source of friction closes.
• When you acquire the next business, the system adapts to their routing rules instead of forcing them into your workflow.

The Trust Question

Every dispatch manager we talk to asks the same thing: “What if the algorithm makes a bad call?”

It’s the right question. And the answer is: you override it. The system recommends. You decide. Every route plan is reviewable. Every assignment is adjustable. You’re not handing control to a black box, you're giving your best dispatcher a tool that sees the entire board instead of just the next three moves.

The first few weeks, you run it alongside your current process. You compare what the algorithm suggests versus what your team would have done. What we consistently see is that the algorithm catches pairings and efficiency the manual process missed, not the other way around. Because it’s processing every variable across every load simultaneously. No human brain does that at 250 loads a day.

Trust is earned. That’s exactly how it should work.

Who This Is For

If you’re running a private fleet with complex routing requirements multiple trailer types, cross contamination rules, flexible and rigid delivery windows, intraday re-optimization needs and the big platforms are either too expensive, too rigid, or too bloated for what you actually need, this is the approach worth looking at.

If your dispatchers are making 250 routing decisions a day based on memory and habit, and you know there’s margin hiding in better backhaul utilization, this is how you find it.

If you’re growing through acquisitions and need a routing system that adapts to new businesses instead of forcing them into a template, this is how you build for scale without locking yourself into a five year contract with a legacy vendor.

Where to Go From Here

We talk about fleet optimization and custom routing systems on our podcast and YouTube channel real conversations with ops leaders about what works in the field and what doesn’t. If you want to see what an optimized routing workflow looks like mapped to your actual operations, we’re happy to walk through it.

Book a call with the ScaleLabs team and bring your messiest routing day. We’ll show you exactly where the margin is hiding.