Your Best Month Is Your Leakiest Month

Somewhere in your books is a record month. Harvest, if you haul grain. Late summer, if you're paving. The month after a hail storm, if you run field crews. Volume tripled, everyone worked weekends, and the invoice total came in higher than any month before it.

Here's the claim worth arguing about: that month, the one you're proudest of, is almost certainly the month you leaked the most money. Not because your people got careless, but because leakage isn't spread evenly across the year. It concentrates exactly where the volume does. If you're going to audit one month of tickets, audit your best one.

Volume is variable. Your back office isn't.

When a quarry's daily load count triples, the scale keeps up fine. Scales don't get tired. What doesn't triple is the billing office. The same two people who keyed 400 tickets a week in March are keying 1,200 a week in October, plus the phone is ringing more, plus half the drivers are seasonal hires who've never handled your paperwork before.

Every failure mode you already have gets worse under load, and none of them get worse in proportion. They get worse all at once.

Start with the exception pile, which grows fastest exactly when nobody has time to work it. A ticket with a missing PO takes two minutes to resolve in a slow week. In a peak week it goes in the pile, because the clerk has 300 clean tickets behind it. The pile doesn't get worked until the surge ends, and by then the job's closed, the seasonal driver is gone, and nobody remembers whether load 40312 went to the north field or the south one.

Miskeys climb with pace. A clerk hours deep into a stack types 26.4 tons as 24.6 and moves on. Nothing flags it. It's a plausible number on a plausible ticket.

And the handoffs multiply. Peak season means borrowed trucks, brokered loads, a customer's own driver hauling from your pit. Every one of those is a ticket copy in the hands of someone who doesn't work for you and won't be here next month.

None of this is news to your billing team. Ask them which weeks they dread and they'll name your revenue peaks without hesitating.

flowchart TD
  A["Peak week: ticket volume triples"] --> B["Same two clerks key the pile"]
  B --> C{"Ticket has a problem?"}
  C -- no --> D["Keyed and billed"]
  C -- yes --> E["Set aside for later"]
  E --> F["Job closes, seasonal driver leaves"]
  F --> G["Billed late, billed short, or never billed"]

The record hides the miss

Now the second half of the problem, and it's the half owners tend to resist. The month you leak the most is also the month you're least able to see it.

Run the numbers on a hypothetical. Say your operation does $250K in a normal month and leakage runs 1% in slow periods, call it $2,500 slipping through on exceptions and miskeys. In your surge month, revenue hits $600K, and under load your leakage rate triples to 3%. That's $18,000 gone in one month, more than half a year of normal-rate leakage in four weeks.

But what does the owner see? A $600K month. The best top line in company history. Nobody stares at a record and asks why it wasn't $618K. Every instinct you've built for spotting trouble, comparing this month to last month, watching for revenue dips, checking whether billing kept pace with work, all of it runs on variance. In a surge month, everything is up. The miss is invisible because it's a smaller piece of a bigger number.

Compare that to a slow February. Volume is low, the clerks have time, exceptions get chased the same day they appear, and fewer tickets leak in the first place. Leakage never shows up as a shortfall against an expected number in any month, because nobody has an expected number for loads that were never billed. But in February the process that prevents it is running at full attention. You have your tightest controls when they matter least and your loosest when they matter most.

A reasonable operator will push back here: we've run twenty harvests, we always catch up in November, our customers would tell us if invoices were wrong. Take the last one first. Customers call about overcharges. Nobody calls to say a load is missing from their invoice. The catch-up work in November finds the exception pile that survived, but it can't find the driver copy that never reached the office, because there's nothing in the office to find. And "we've always done it this way" is precisely what you'd say if the leak had been running for twenty years.

Audit the peak, not the average

The practical consequence is about where you point your attention.

Most operators who decide to check their billing pull a recent, typical month, on the theory that typical is representative. For leakage, it isn't. Pull your biggest month instead, and start with counts before you read a single ticket. How many loads does the scale system or dispatch log show for the peak week? How many ticket numbers exist in sequence? How many billed lines went out? Three numbers, one afternoon. If they match, you've earned real confidence, because you checked your process at its breaking point. If they don't match, the gap is specific: these ticket numbers, this week, this customer, still unbilled.

Then look at your calendar. If your surge is seasonal, you know when the next one starts. The weeks before it are the one window where tightening intake actually holds: get tickets captured digitally at the point of handoff instead of riding around in trucks, agree on what happens to an exception ticket the day it appears rather than the month after, and make sure the seasonal hires' first day covers what to do with paper. Every ticket that gets captured and checked before the pile forms is one that can't quietly miss the cutoff.

That's the work Termco does, digitizing tickets as they're created and flagging the incomplete and non-compliant ones while the job is still open and the driver still answers the phone. But the audit of your last peak month costs you nothing and needs no software, and it's the fastest way to find out whether your best month was as good as the top line says.

If you'd rather have a second set of eyes on it, send 10-20 sample tickets from your busiest month to hi@termco.ai, and we'll send back what we find within 48 hours.