“But we have a service level agreement for the billing runs. The vendor has hit 99.8% accuracy for three quarters straight.”

“Great. So why are there two hundred people in the lobby of our workflow system screaming for a human to look at their accounts?”

“That’s different. Those are disputes. Modifications. In-life changes. That’s judgment-based work. We keep that in-house because you can’t really outsource ‘it depends’.”

It is 4:22 PM. I started a diet at 4:00 PM, and I can already feel my blood sugar performing a slow, dramatic swan dive into a pool of existential dread. I am currently staring at a cold cup of coffee and wondering if a spoonful of sugar constitutes a “relapse” or merely a “strategic adjustment.”

This is how most people approach their back-office operations: they call a massive failure a “strategic adjustment” because they’ve outsourced the part of the business that was actually easy to solve, while clutching the broken parts to their chest like a prized heirloom.

A Tale of Two Operations

The scene is almost always the same. You walk into the operations center of a mid-sized lender, and you see two very different worlds. On one side, there’s a sleek monthly report from a Business Process Outsourcing (BPO) firm. It’s full of green checkmarks and “transactions processed” metrics.

They’ve keyed in 50,000 payments. They’ve generated 50,000 statements. They’ve mailed 50,000 notices. It’s efficient. It’s cheap. It’s also, quite frankly, a task that a well-configured toaster could perform if it had an API connection.

Then you look at the other side of the room. This is the “In-House Judgment Team.” There are four of them, and they look like they haven’t seen sunlight since the Obama administration. They are buried under a backlog of 191 items.

These aren’t “transactions.” These are “situations.” An unapplied cash situation where a lessee sent a check for $4,512.18, but the system expected $4,400.00 plus a tax rate that changed three weeks ago but wasn’t updated in the static data. A dispute over a piece of equipment that was “returned” but the carrier lost the bill of lading. A mid-term modification where a fleet manager wants to swap ten tractors for twelve trailers and keep the same monthly payment.

The H-Point of Impact

The firm has outsourced the easy half and kept the half that requires judgment. On paper, this sounds like a sophisticated management decision. You keep the high-value, high-context work close to the vest. But the economics are a disaster.

Sarah C.-P., a woman who spent most of her career as a car crash test coordinator, once explained to me that the most expensive part of a crash isn’t the impact itself; it’s the variables you didn’t account for in the rig setup.

– Sarah C.-P., Crash Test Coordinator

If the sled hits the wall at 35 miles per hour, that’s data. You can measure that. But if the dummy’s “H-point”-the pivot point of the hip-is misaligned by half an inch, the entire multi-million dollar test is garbage. You’ve spent all your money measuring the “obvious” (the speed of the hit) while ignoring the “nuance” (the pelvic orientation) that actually determines if the human survives.

In equipment finance, statement production and payment keying are the “speed of the hit.” They are observable, specifiable, and boring. Because they are specifiable, they are easy to write into a contract. You can hold a vendor’s feet to the fire over a late billing run.

But because they are easy to specify, they are also the easiest things to automate. Meanwhile, you’ve retained the “H-point” work-the exceptions-and you’ve left your best people to handle them with nothing but a spreadsheet and a prayer.

The Economics of Friction

The tragedy is that “judgment” in the back office is often just a polite word for “fixing errors the system allowed to happen.” When unapplied cash sits in a suspense account for forty-five days, it isn’t because the staff lacks judgment. It’s because the system lacked the flexibility to suggest a match.

When a dispute over a casualty loss drags on, it’s because the collateral records and the insurance data are living in different zip codes. We outsource the things we can describe and keep the things we can’t. But the things we can’t describe are usually the things that are costing us the most.

This is the central paradox of modern lending operations. We treat human capacity as a bucket for overflow rather than a tool for precision. If your staff is spending 80% of their time manually correcting tax jurisdictions or chasing down missing serial numbers, they aren’t “exercising judgment.” They are acting as a human bridge between two islands of bad data.

The real shift happens when you realize that the routine isn’t a commodity to be shipped overseas; it’s a burden to be eliminated through architecture. If you use a modern, API-first platform like

equipment finance software,

the goal isn’t just to “process” the lease; it’s to ensure the data is so clean that the “exceptions” actually become exceptional.

The Salt-Dusted Almond Test

I’m currently wondering if my diet allows for a single almond. A single, salt-dusted almond. I am weighing the “judgment” of this decision against the “specification” of the diet plan. The plan says no snacks. My judgment says I might faint and hit my head on the radiator.

This is a high-stakes exception. If I had a better “system”-say, a meal plan that didn’t rely on my own failing willpower at 4:30 PM-I wouldn’t be in this position.

Businesses find themselves in the same spot. They rely on “human willpower” to bridge the gaps in their legacy servicing systems. They call it “customer service” or “operational expertise,” but it’s really just a tax on inefficiency.

Think about the 191 items in that backlog. If you dig into them, you’ll find that 150 of them shouldn’t exist. They are the result of a system that is too rigid to handle a partial payment or too dumb to recognize a change in a sales tax rate. By keeping these “exceptions” in-house, you are effectively subsidizing your own technical debt with the life-force of your operations team.

The Factory vs. The Clinic

When you automate the routine, you don’t just “save headcount.” You change the nature of the work. You move from a “factory” model, where people are judged by how many widgets they move, to a “clinic” model, where people are judged by how many problems they solve.

The irony is that the vendors we outsource the “easy” work to are often using the same antiquated tech we are. They just have cheaper labor to throw at the screen. This doesn’t solve the problem; it just makes the problem quieter. A quiet problem is still a problem, especially when it’s sitting in a suspense account or a delinquency bucket.

“I once worked with a lender who was incredibly proud of their ‘exceptions team.’ They were the ‘Navy SEALs’ of the back office. They could fix anything.”

But when we actually looked at the data, we realized these “SEALs” were spending six hours a day re-typing data from one screen to another because the two systems didn’t talk. They weren’t SEALs; they were highly paid data entry clerks with fancy titles.

We have to stop valuing “judgment” when what we’re really asking for is “manual workarounds.” Real judgment is the ability to look at a customer’s deteriorating credit and decide to restructure a deal before it goes to collections. Real judgment is figuring out how to package a new green-energy financing product that doesn’t fit the traditional lease mold. You can’t do that when you’re 191 items behind on unapplied cash.

Liberating the Un-Specifiable

The boundary decisions we make-what to keep, what to send away-are usually based on how easy it is to write the contract. It’s easy to write a contract for “bill 10,000 people.” It’s hard to write a contract for “make our customers feel like we understand their business.”

But the “hard to write” part is where the value lives. If you automate the specifiable, you liberate the humans to handle the un-specifiable. You stop treating your staff as a buffer for your software’s inadequacies and start treating them as the competitive advantage they are supposed to be.

My coffee is now cold and my stomach is making a sound like a distant thunderstorm. The diet continues, mostly out of spite. I’ve realized that the “routine” of my day-the checking of boxes, the answering of emails-is the easy stuff. The “judgment”-deciding what actually matters when you’re hungry and tired-is the part that needs the most support.

If you are still running your portfolio on a system that requires a “judgment team” to handle basic data corrections, you aren’t just behind the curve. You are paying a premium for the privilege of staying there. The goal isn’t to outsource the easy half; it’s to automate the easy half so thoroughly that the hard half actually becomes manageable.