This is where most of the work happens. Every week you count, you order, you receive, you cost, you fix what staff broke. We are opinionated about how to run that loop — what to count and when, how to read variance without spiraling, what to do when a vendor changes a pack size on you. Skim the sub-sections in any order; the loop is a wheel, not a ladder.
Chapter II
II
Run the loop
Count your walk-in in 10 minutes by voice
Counting the walk-in shouldn't take an hour. If yours does, you're counting the wrong things in the wrong order — or you're writing instead of talking.
Our take
Most independent operators count their walk-in once a week and hate every minute of it. We think weekly is right and one hour is wrong. A 2-door walk-in with ~80 SKUs should take 10 minutes by voice, every Sunday night, in the same order, every time. The point of the count isn't to know every can of tomato to the unit. It's to catch the 15 items that move enough money to matter — and to do that often enough that variance stays small.
How to count your walk-in
Start at the door, top shelf, left. Always. Voice counting only works if your route is predictable — the app expects items in the order you've set up your count sheet. (Screenshot: walk-in entry, numbered shelf overlay 1–5.)
Tap the mic, say the item and the count. "Romaine, two cases and three heads." The app handles the conversion to your purchase unit. You don't have to think in eaches.
Move shelf by shelf without backtracking. If you missed an item, keep going — you can append it at the end. Looping back doubles your count time and your error rate.
Weigh proteins, eyeball produce, count cases. Anything over $40/lb gets the scale. Anything under, eyeball. The math (see below) says perfection on cheap items is wasted minutes.
End at the door, bottom shelf, right. Confirm the count when prompted. (Screenshot: in-app confirmation screen.)
Don't recount when the variance flag pops. The app will flag romaine if you said "two cases" and last week's depletion suggests you should have 0.5. Look at the romaine, then accept or correct. The flag is information, not an accusation.
Why this works
Three things make voice counting faster than paper or tap-counting:
Your hands stay free. A walk-in count is also a freshness audit — you should be turning packages to check dates. With a clipboard, you put it down and pick it up 30 times. With voice, you don't.
The order is fixed. Inventory accuracy comes from counting the same things in the same order every time. Voice forces consistency because the app expects the route. Clipboards let you skip around; skipping around is how items get missed.
Conversions happen at the edge, not in your head. "Two cases of romaine" is what your brain knows. "12 heads" is what your invoice knows. The app translates. You stop doing mental arithmetic at 9pm.
The math
Don't do this
What good looks like
A completed Sunday walk-in count for a 38-seat cafe usually lands at 78–84 items, 9–11 minutes, 3–6 variance flags. Flags resolve in under 60 seconds each. The whole loop, including the freezer, is done before 10pm and ready to inform the Tuesday produce order. (Screenshot: completed count log with realistic sample numbers — 82 items, 10:14 elapsed, 4 flags resolved.)
If your number is double that — 20 minutes, 15 flags — your count sheet is out of date. Walk it once with a manager and prune.
Where this connects
Set your weekly loop — Why Sunday night is the right slot, and what else lives in it.