By Matt Wampler, CEO of ClearCOGS
Quick Answer: The final batch of the day should be sized to the demand remaining in the hours you have left, not to the comfort of the person making it. Most kitchens have a reasonable morning number and no number at all for the last call, so a manager estimates it at the busiest moment of the evening with the least information available. That one decision drives a large share of end-of-day waste, because anything left when the doors close on a non-carryover item is a total loss, including the packaging it was portioned into.
The morning number is usually fine
Here is something worth saying plainly, because it surprises people: most well-run operations have already solved the first prep decision.
They have a build-to list, usually by day of week, often derived from product mix at some point and reviewed periodically. It says make this much to get through lunch. It is a little generous on purpose. A kitchen running that system will typically finish the lunch period with a small amount left, which is exactly what it was designed to do.
That is a functioning system. It is not guesswork, and the people running it are frequently very good at it after a decade of practice.
The trouble is that the system covers the part of the day that is easiest to predict, and goes quiet for the part that is hardest.
Where the decision actually gets made
Walk the same kitchen at eight in the evening.
The afternoon has been unpredictable. The dinner rush may have landed early or not at all. Someone now has to decide whether to drop another load of protein, fire another batch of a hot side, or stop producing and start cleaning.
That decision has three properties that make it uniquely expensive.
It is irreversible in one direction. Make too much at ten in the morning and most of it still sells. Make too much at eight at night and there is no later. For a non-carryover item, the leftover is worth nothing at all the following morning.
It is made by whoever happens to be on shift. The general manager is not there every night. The closing manager may be the least experienced person who will touch a production decision all day, and they are making the most consequential one.
It is made under the strongest possible pull toward over-producing. A rush that feels busy at 7:45 makes “drop another load” feel obviously correct. Fifteen minutes later the cars stop coming, and a large quantity of prepared product has nowhere to go.
So the brand has invested real effort in the number that is easy, and left the number that is hard to the instinct of the person with the least support.
Why the end of the day is genuinely harder
It is worth being fair about this, because the instinct to blame the closing manager is strong and mostly wrong.
Two things make the final hours harder than the rest of the day in a measurable way.
First, the volume is lower, and lower counts are inherently noisier. Three orders in a half hour is not a smaller version of thirty. The natural variation is a much larger share of the total, so even a good estimate feels unreliable.
Second, product that cannot be held overnight belongs to a different class of inventory problem than product that can. The operations research literature treats fixed-life perishable goods as a separate and substantially harder category from ordinary stock, with its own body of work on ordering policies specifically because the standard approaches do not transfer (Nahmias, Perishable Inventory Theory: A Review, Operations Research, 1982). A case of frozen fries and a pan of gravy are not the same decision, even though they sit on the same prep sheet.
Put those together and the last call of the night is the noisiest estimate, on the least forgiving item, made by the least supported person, at the moment they are busiest.
What is actually being thrown away
One more thing operators underweight, and it matters for the arithmetic.
On prepared hot items, the raw ingredient is frequently the cheap part. What makes the discard expensive is everything added to it: the labor to produce the batch, and in a portioned concept, the container and the lid it went into. Throw away a dozen portions of a side dish and you are not out a dozen servings of the ingredient. You are out the ingredient, the prep time, and twelve pieces of packaging.
That is why end-of-day waste on inexpensive items adds up faster than the food cost line suggests, and why it hides from a review that only looks at ingredient prices.
What a better system looks like
The fix is not to tell managers to be more careful. It is to give the last decision the same support the first one already has.
- Give the final window its own number. If your build-to list stops at the lunch number and a vague second batch, it is missing the decision that costs the most. The last two hours deserve their own quantity, by day of week, the same way lunch does.
- Set an explicit end-state target. Most operators already have one in their heads, usually some version of running out shortly before close. Write it down. “Zero gravy at 8:45” is a target a manager can hit. “Don’t run out, but don’t waste” is not.
- Decide per item, not per kitchen. An item with a close substitute on the menu and high prep labor should be run tight at the end of the night. A signature item that brings people in should not. Those are different answers and most prep sheets give them the same one.
- Make the number available to whoever is actually closing. If the guidance lives with the general manager, it does not exist on the nights that matter. It needs to be in front of the person on shift.
- Record the outcome, briefly. What was made in the last batch and what was left at close. Four weeks of that turns an argument into a pattern, and it is the only way to know whether the target is right.
Frequently Asked Questions
How much should a kitchen prep in the last batch of the day?
Enough to cover expected demand for the remaining open hours, adjusted for how costly it is to run out of that specific item. For an inexpensive item with a substitute, that usually means planning to finish close to empty. For a signature item, it means carrying a deliberate cushion.
Should a restaurant plan to run out before closing?
On non-carryover items, often yes. If the product has no value the next morning, finishing the night close to zero is usually cheaper than holding a buffer that gets discarded. That should be a deliberate target rather than an accident.
Why does most waste show up at the end of the day?
Because that is where the irreversible decisions are. Earlier over-production is recoverable through later sales. Late over-production has no remaining selling time, so the entire error becomes waste.
Who should decide the final batch?
Whoever is on shift has to execute it, but they should not have to originate the number. The quantity should come from the same system that produces the morning number, with the closing manager adjusting for what they can see in the dining room.
Does this apply to items we hold over?
Much less. If product carries to the next day with acceptable quality, a late over-production is a timing error rather than a loss. The sharpest version of this problem is specific to items with no carryover.
Most brands have built a system for the production decision that is easiest to get right and left the hardest one to instinct. The morning number gets a list, a review, and years of refinement. The eight o’clock call gets a shrug and whoever is on the schedule.
If you want to find end-of-day waste, do not start with the recipe or the ingredient price. Start by asking who decides the last batch, what number they are working from, and whether anyone has ever written down what the kitchen should look like at close.
If nobody can answer that, you have found the gap.
Sources
- Nahmias, Steven. Perishable Inventory Theory: A Review. Operations Research, 30(4), 680–708. August 1982. doi.org
