How to reconcile van stock at the end of the day, not the end of the week

A bakery and dairy distributor running seven vans out of one depot closed the books every Friday. Not every day — every Friday, because that was the only time the office had a free block to sit down with a week's worth of paper delivery notes, a week's worth of cash bags, and a spreadsheet that tried to hold both. By Friday afternoon, someone was always short. The question was never if there was a discrepancy. It was whose fault it was, and by then nobody could remember.

The reconciliation nobody could actually do

The theory of van stock reconciliation is simple: what left the depot on a van should equal what was sold, plus what came back unsold, plus what was returned as damaged or expired, plus or minus any documented adjustment. Three columns that have to add up — load, disposition, cash — for every van, every day. In practice, the bakery distributor's version of this was a driver handing in a stack of carbon-copy delivery notes and a cash bag, once a week, for someone in the office to key in by hand.

By the time that happened, a driver couldn't say why Tuesday's bread count was four short — Tuesday was four days ago, and he'd loaded and unloaded the van a dozen times since. A customer who'd disputed a delivery on Wednesday had already been re-invoiced by memory, not by record. A weekend of high humidity had spoiled a tray of dairy that got written off verbally, with no line anywhere connecting the loss to the load it came from. None of these were large on their own. Stacked across seven vans and five days, they were a number the owner reviewed every Friday with a mix of dread and resignation, because the format of the review — paper against a spreadsheet, a week late — made the number impossible to actually explain, only accept.

Why weekly reconciliation always loses information

The core problem wasn't dishonesty or carelessness. It was that reconciliation was being attempted at the wrong distance from the event. A driver can tell you exactly what happened to a case of yogurt an hour after it happened — a customer refused it, it went back on the van, it's sitting in the third bay from the left. Ask him five days later and it's just "a discrepancy," because the specific memory that would explain it is gone, and the paper trail was never designed to preserve it. Weekly or even daily-but-manual reconciliation doesn't fail because people are bad at math. It fails because the information needed to close the loop has a shelf life measured in hours, and the process runs on a cycle measured in days.

This is also why the instinct to add more oversight — a supervisor double-checking the sheets, a stricter sign-off — rarely helps. More scrutiny on stale data still produces stale answers. The fix has to happen upstream of the paperwork, at the moment stock actually changes hands.

The turning point: reconciling per route, per day, at the point of the transaction

The change that worked wasn't a new spreadsheet template. It was moving the three-way match — load, disposition, cash — onto the same device the driver already used to record every sale, and running it the moment the van returned instead of at week's end.

Each morning's load-out is recorded as a count against the route plan, not a line on a clipboard. Every sale, return, and damage write-off during the day is logged against that same load, in the unit it actually moved in, with a reason code if it isn't a straight sale. When the van pulls back into the depot, the app already knows what should be left: opening load, minus sales, minus returns, minus write-offs, equals expected remaining stock and expected cash. The driver counts what's physically on the van and what's in the bag, and the system shows the variance — if any — immediately, while the driver is standing right there and can still explain it.

That single change collapsed a week of guesswork into a same-day conversation. A four-case shortfall stopped being an unsolvable mystery and became a two-minute conversation about a delivery that got shorted at a specific stop that morning. Real shrinkage, when it existed, showed up as a pattern against one van or one SKU within days instead of being buried in a monthly average. And because the numbers closed daily, the office stopped spending Friday afternoons refereeing disputes and started spending it on the two or three genuine exceptions that needed a manager's attention.

How to actually reconcile van stock

• Record the load-out as a counted transaction, not a handwritten tally — it's the baseline every later number gets compared against.
• Log every disposition at the moment it happens: a sale, a return, a damage write-off, a customer dispute — each with a reason code, entered from the same device as the sale.
• Run the three-way match daily, per van — loaded stock vs. sold vs. returned/written-off vs. cash collected — the day it happens, not batched at week's end.
• Surface the variance to the driver before they leave the depot, while the specific stop or swap is still in recent memory.
• Track variance trends by van and by SKU, not just the day's total — a repeat pattern is a process fix; a one-off is just a one-off.

iotoms builds this reconciliation into the same field app the rep already carries: the load-out, every sale, return and adjustment, and the end-of-day cash count all flow into one route settlement that closes to the cent before the driver goes home — so the office sees the day's exceptions the same afternoon, not the following Friday.