Why Restaurants Run Out of Ingredients (and How to Stop 86ing Items Mid Service)

Restaurants rarely run out of ingredients because of bad luck. When something gets 86'd during service, the cause can almost always be traced back to the last order: the count was old or wrong, the par was built for an average day instead of the days the order had to cover, the order missed the vendor's cutoff, a short on the truck went unnoticed, or someone confused cases with eaches. Sometimes the raw product is in the walk in and the line still runs out, because prep didn't keep up.

To stop it, write down every 86 for a few weeks, match each one to its cause, and fix that cause. Ordering more of everything won't do it. Most kitchens find that the same two or three causes account for most of their 86s.

In this guide

• The eight usual reasons a kitchen runs out
• Diagnosis table: symptom, likely cause, fix
• Worked example: burger patties that keep running out on Sunday
• Start an 86 log
• What line cooks and chefs can do during service
• FAQ
• Keeping counts, pars, and cutoffs in one place

The eight usual reasons a kitchen runs out

1. The count was missed, stale, or guessed

Every order is only as good as the number it starts from. If the count happened the night before and the kitchen ran a busy brunch the next morning, or if someone eyeballed the walk in instead of counting, the order is built on a number that's already wrong.

Watch for these:

• Counting after the order was placed, or skipping the count on a busy day and copying last week's number.
• Counting the main walk in but forgetting the reach in on the line, the freezer, or the dry storage upstairs.
• Counting before a delivery on one day and after it on another, so the numbers don't mean the same thing from week to week.

The fix is a count timed to each order, on a sheet that follows the shelves. For high risk items, a short daily count is often enough. Our guide to doing a restaurant inventory count covers full counts versus short daily counts and how to time them to each order.

2. The par is too low for the days it has to cover

A par is supposed to carry you from the moment you count until the delivery after the one you're ordering. Pars go wrong in a few predictable ways:

• Built from an average day. If Friday and Saturday sell twice what Tuesday does, a par built on average daily usage will come up short on the weekend order.
• One par for every order day. A Thursday order that has to last through the weekend needs a bigger par than a Monday order that only covers the slow part of the week.
• No safety stock. A par with zero buffer runs out any time a night is busier than average, and that happens all the time.
• Set years ago. The menu changed, the item moved from a special to a regular dish, and nobody updated the number.

We won't repeat the full method here. How to set par levels for a restaurant walks through it step by step, including separate pars for uneven delivery gaps.

3. The order missed the cutoff

This one hurts more than people expect. If a vendor delivers twice a week and you miss Thursday's cutoff, you don't wait one extra day. You wait for the next truck, which might be Tuesday. The worked example below shows what that does to a weekend.

Cutoffs get missed because they live in someone's head, because they differ by vendor (noon for one, 4 PM for another, two days ahead for a specialty purveyor), or because the person who usually orders is off. Write every vendor's cutoff and delivery days in one place and work your count times backward from them. Our vendor ordering schedule guide shows how to build that.

It also helps to confirm the order actually went out. A text that never sent or a portal cart that was never submitted looks exactly like a missed cutoff when the truck shows up without it. A simple order log with a "submitted" status catches this. See how to track restaurant purchase orders.

4. Lead time and delivery gaps weren't counted

Lead time is the gap between placing an order and having the product on your shelf. For a vendor with a noon cutoff and next morning delivery, that's about a day. For a specialty fish or meat purveyor, it could be two or three days, and some products need to be ordered further ahead than that.

The mistake is planning only for the gap between deliveries and forgetting that the kitchen keeps cooking while the order is in transit. If you count Thursday morning for a Friday truck, Thursday's service comes out of what's already on the shelf. Your coverage window is:

Days to cover = lead time days + days until the following delivery

5. Shorts and subs weren't caught at the back door

If the truck brings two cases instead of three, or a different product in a different pack size, and nobody notices until a cook opens the box during service, you've lost the time you needed to fix it. Check every delivery against the purchase order, not just the invoice, and flag anything missing while the driver is still there. The reconciliation section of our purchase order tracking guide shows how to compare what you ordered with what arrived and what you were billed.

6. Somebody confused cases, packs, and eaches

Unit mix ups cause some of the strangest run outs, because the order looks reasonable on paper.

Illustrative example: a kitchen counts short rib in packages and buys it in cases of 6 packages. The par is 24 packages. The closer counts 2 packages and writes "2" on the sheet with no unit. The next morning, the person ordering reads it as 2 cases, or 12 packages.

What they think:  24 − 12 = 12 packages short → 2 cases
What's true:      24 − 2  = 22 packages short → 22 ÷ 6 = 3.67 → 4 cases

On hand after the delivery:  2 + (2 × 6) = 14 packages against a par of 24

That's 10 packages short of par, and nobody did anything obviously wrong. Put a unit next to every count and keep the case size on the sheet. Our guide to restaurant inventory units covers count units, order units, and conversions.

7. Prep didn't keep up, even though the product was there

Sometimes the walk in is full and the line still 86s the dish. The raw product isn't the problem. The prepped version is. Braised short rib, stocks, doughs, and anything that has to marinate, cure, or thaw can't be made in the middle of service.

For these items, the lead time that matters is the prep time. If short rib braises for most of an afternoon and needs to cool, Saturday's short rib has to be cooked Friday. That means the prep par for Friday's braise is set by Saturday's sales, not Friday's.

What helps:

• Set prep pars by day of the week, just like order pars. Our guide to writing a restaurant prep list shows how to work them out from covers and subtract what's already made.
• For long prep items, decide the day before how much to start, based on the next day's expected covers.
• Have cooks call it when the prepped item is getting low, not when it's gone (more on that below).

8. Something unusual happened: a holiday, event, or big party

A nearby concert, a holiday weekend, a heat wave on a patio, or a 40 person party booked on a Tuesday can blow through a par that works fine the rest of the year. A normal par isn't supposed to cover these. Plan for them separately:

• Check reservations, private events, and the local calendar before each order, not just on the day.
• Add the extra to the order that has to cover the event, and adjust the order after it so you don't end up overstocked.
• Ask your reps early whether holidays change their delivery days or cutoffs. A delivery day that disappears is the same as a missed cutoff. Our holiday ordering guide walks through a Thanksgiving week where that happens.

A quick word on vendor outages

Sometimes the vendor simply doesn't have it. That's a real cause, but it's less common than the ones above, and it's the one kitchens tend to blame first. If one item gets shorted again and again, keep a backup source written on your order guide and consider a little more safety stock on that item while you sort it out with the rep.

Diagnosis table: symptom, likely cause, fix

Use this when you look over your 86 log. Find the line that matches what happened, then start with the fix in the last column.

Worked example: burger patties that keep running out on Sunday

This is an illustrative example. The restaurant, vendor schedule, usage, and quantities are made up to show the method. They aren't typical numbers for any concept.

The setup

• The restaurant is open seven days a week.
• Burger patties are counted in patties and ordered in cases of 40.
• The meat vendor delivers Tuesday and Friday mornings, before lunch. The cutoff is noon the day before.
• The kitchen manager counts and orders on Monday and Thursday mornings.
• The patties have a use by date that covers the window. (If yours don't, shelf life caps the par. Check that before copying this math.)

The 86 log shows patties went out on two of the last four Sunday nights, and once on a Monday.

Usage by day of the week

The manager pulls the last four weeks of counts and invoices and works out average usage for each day:

What the old par missed

The par on the count sheet was set from the average day:

Average day:        440 ÷ 7 = 62.9 → about 63 patties
Old Thursday par:   63 × 5 days = 315 patties, no safety stock

The five days are right. The Thursday count has to cover Thursday (lead time: the truck isn't here until Friday) plus Friday through Monday (until the Tuesday truck):

Days to cover = 1 day lead time + 4 days until the following delivery = 5 days

The problem is which five days. Thursday through Monday includes the two busiest days of the week:

Actual Thu to Mon usage:  55 + 90 + 110 + 60 + 40 = 355 patties
Old par:               315 patties
Gap:                   355 − 315 = 40 patties short, before anyone has a busy night

Here's last Thursday. The count was 95 patties.

Order:     315 − 95 = 220 patties → 220 ÷ 40 = 5.5 cases

The manager rounded down to 5 cases (200 patties) to save walk in space. Here's how the weekend went:

Thursday end:      95 − 55             = 40
Friday morning:    40 + 200 (delivery) = 240
Friday end:        240 − 90            = 150
Saturday end:      150 − 110           = 40
Sunday:            40 on hand, 60 needed → out partway through Sunday dinner

Rounding up to 6 cases wouldn't have saved it either. That gives 335 patties against 355 needed, so the kitchen runs out Monday night instead.

The fix: par from the actual days, plus a buffer

The busiest Saturday in the last four weeks used 130 patties, 20 more than the 110 average. The manager uses that as the safety stock. It's a judgment call, not a formula.

New Thursday par:  355 (Thu to Mon usage) + 20 (safety stock) = 375 patties
Order:             375 − 95 = 280 patties → 280 ÷ 40 = 7 cases exactly

Tracing the same weekend with 7 cases:

Thursday end:      95 − 55             = 40
Friday morning:    40 + 280 (delivery) = 320
Friday end:        320 − 90            = 230
Saturday end:      230 − 110           = 120
Sunday end:        120 − 60            = 60
Monday end:        60 − 40             = 20 left when the Tuesday truck arrives

The 20 left on Tuesday morning is the safety stock, still there because nothing unusual happened. On a weekend where Saturday runs 20 over, it gets used, and the kitchen still makes it to the truck.

The Monday order covers a slower stretch, so it gets its own par:

Monday window:     1 day lead time (Mon) + Tue, Wed, Thu = 4 days
Usage:             40 + 40 + 45 + 55 = 180 patties
New Monday par:    180 + 20 = 200 patties

Using 315 on Mondays had been carrying more patties than needed through the slow part of the week, while the weekend order came up short.

What a missed cutoff would have done

Say the Thursday order goes in at 1 PM, an hour after the cutoff. The next truck is Tuesday.

Thursday end:      95 − 55 = 40
Friday:            40 on hand, 90 needed → out during Friday dinner

That's three and a half days without patties. With a twice weekly vendor, missing one cutoff costs you a whole delivery gap, not a day. If that happens, call a backup vendor or plan a store run right away, on Thursday afternoon, not Friday at 7 PM.

Start an 86 log

You can't fix what nobody remembers on Monday. An 86 log is a running list of every item that ran out, kept somewhere the whole team can reach: a clipboard by the pass, a page in the manager's binder, or a tab in your purchasing spreadsheet.

What to record

Illustrative entries.

Keep it to what someone can fill in quickly during service: the item, the time, and what was left. The manager can fill in the cause and fix the next morning, while people still remember what happened.

The "raw or prepped" column matters more than it looks. It tells you right away whether to look at the order or at the prep list.

Review it once a week

Before the biggest order of the week, look at the last seven days:

1. Group by item. Anything that ran out twice is a pattern, not a fluke.
2. Group by cause. If most entries say "count" or "cutoff," the fix is a routine, not a par.
3. Fix one thing. Change a par, move a count time, add an item to the order guide, or add a cutoff to the schedule.
4. Check last week's fixes. If the item hasn't run out since, the fix worked. If it has, look again at the cause.

Also note items you threw out or that went bad before you used them. If a par goes up to stop 86s, waste is how you find out it went up too far.

What line cooks and chefs can do during service

Most 86s are decided days earlier, but how service handles the last few portions still matters.

• Call it early. "Six patties left" at 7 PM gives expo and the floor time to steer guests. "We're out" with tickets on the rail doesn't.
• Check every storage spot before calling the 86. The reach in, the walk in, the freezer, and the prep shelf. A surprising number of 86s get called while a case sits in another cooler.
• Tell one person. The chef or manager decides on a sub or a sellout. Don't quietly swap in a different product, especially if it changes allergens or the portion.
• Write it on the log. Item, time, and what was left. Ten seconds.
• Flag prep that's running behind. If the braise or the dough won't be ready in time, say so at the start of service, not when it runs out.

FAQ

Is it ever fine to run out of something?

Yes. Plenty of kitchens prep a set number of a special or a limited dish and sell out on purpose, because it keeps waste down and the product fresh. The difference is that a planned sellout is a decision you made ahead of time. An unplanned 86 on a core menu item is what this guide is about.

How much safety stock should I keep?

There isn't one right number. Size it by how much the item's usage swings, how reliable the vendor is, and how bad it is to run out. A signature protein deserves a bigger buffer than a garnish, and perishables should get smaller buffers than dry goods. In the example above, the buffer was the difference between the busiest Saturday and the average Saturday. Our par level guide covers this in more detail.

Why do we run out when we order the same amount every week?

Because your usage doesn't stay the same every week. Standing orders don't account for what's already on the shelf, for a busy weekend, or for a menu change. Counting before each order and ordering up to a par adjusts the order automatically.

Should we just order more of everything?

No. It ties up cash, crowds the walk in, and turns into waste on anything perishable. Raise pars only on items that keep running out, and only after you've ruled out a count, cutoff, or receiving problem.

How often should we count to avoid running out?

Count the items that drive each order right before you place that order. Many kitchens do a short count of high risk items on order days and a full count weekly or at the end of the accounting period. See how to do a restaurant inventory count for how to split the two.

What's the difference between an 86 and a short?

A short is when a vendor delivers less than you ordered. An 86 is when the kitchen runs out of something during service. A short you don't catch at the back door can turn into an 86 a day or two later, which is why checking the truck against the order matters.

Keeping counts, pars, and cutoffs in one place

Everything in this guide works on paper or in a spreadsheet: a count sheet with units, pars for each order day, a vendor schedule with cutoffs, and an 86 log. Our complete guide to restaurant purchasing and inventory shows how those pieces fit together into one weekly routine.

If you'd rather keep it in one place, Proteus Kitchen stores each inventory item with its current count, a par level, and a minimum, and flags items that are running low or below minimum. Each vendor has its order deadlines and delivery days, and the purchasing dashboard shows which vendors have orders due that day. When you build a purchase order, Proteus shows each item's current count, par, and a suggested quantity, and it keeps a record of submitted orders. Proteus doesn't place orders with vendors for you and doesn't connect to your POS, so you still send each order through the vendor's portal, a text, or an email, and mark it submitted once it's sent.

If your item list, pars, or vendor list are already in a spreadsheet, Free Restaurant Setup covers bringing them in.

Next step: one week of 86s

1. Put an 86 log by the pass today, with the columns above.
2. At the end of the week, give every entry a likely cause from the diagnosis table.
3. Fix the most common cause first, whether it's a par, a count time, a cutoff, or receiving.
4. Repeat next week and see whether that item stays on the menu.