Scrap Rate Formula: How to Calculate It, With an Example

The scrap rate formula is scrap units divided by units started, times 100: the share of what you set out to make that ended up in the bin instead of on a shelf. Scrap is product that can't be sold or fixed: a burnt load, a bag that won't seal and tears when it's reopened, a case dropped on the floor. Product that can be fixed and sold, like an underweight bag that's topped up, is rework, and counted separately. Worked out weekly and split by reason, the scrap rate tells a small maker where money is going in the kitchen, and what each good unit really costs once the scrapped ones are paid for. In the example below, a week of taro chip batches starts 2,400 bags and scraps 58, a scrap rate of 2.4%, and one burnt load is more than half of it. Below is the formula, the example, scrap against yield and rework, the cost, and how to bring it down.

The short version

• Scrap rate = scrap units ÷ units started × 100.
• Scrap can't be sold or fixed; rework is fixed and sold; count them apart.
• Work it out weekly, and split it by reason: that's where the fixes are.
• The cost of scrap is paid by the good units: total cost ÷ good units.
• Yield looks at the same loss from the other side: what came out good against what should have.
• Pick one version of the formula and keep it, so weeks compare.
• TaroStack records what each batch made good, and every loss with its reason. More below.

The formula

Some kitchens divide scrap by good units instead of units started, which gives a slightly higher figure. Either works, if you use the same one every week. "Units" can be finished units, like bags, or weight, like pounds of dough, as long as scrap and the total are in the same unit.

A week of scrap, by reason

Windward Roots, the made-up taro business I use in these examples, fries and packs chips five days a week. In the week of October 5:

Bags started that week: 2,400. Scrap rate: 58 ÷ 2,400 × 100 = 2.4%. Another 22 underweight bags were topped up and sold, a rework rate of 0.9%, so 2,320 bags were good the first time: a first-pass yield of 96.7%.

The split is the useful part. One burnt load is more than half the week's scrap, so the fix is a thermometer check on every load, not a talk about care. The failed seals point at the sealer's temperature or a worn jaw. How a failed check like the burnt load is recorded and dealt with is in nonconforming product form.

Scrap, rework, yield and waste

Yield and scrap are the same loss seen from opposite sides; yield, worked out batch by batch against the recipe, is in batch yield formula. Ingredients that spoil or expire before they're used aren't scrap, because nothing was made from them, but they're still a loss; a log for them is in waste log template.

What scrap costs

The good bags pay for the scrapped ones. If each bag costs $1.95 to make before any scrap, the week's 2,400 bags cost 2,400 × $1.95 = $4,680. Spread over the 2,342 bags that were good in the end:

Cost per good bag    $4,680 ÷ 2,342 = $2.00
Scrap's share        $2.00 − $1.95 = about 5 cents a bag
The week's scrap     58 × $1.95 = $113.10

Five cents a bag doesn't sound like much until it's compared with what a distributor's price leaves; at a distributor price of $2.92 it's about a twentieth of the profit, as worked in distributor margin. Food costs that drift from the recipe for reasons like this one are in food cost variance formula.

How to bring it down

1. Record every scrapped unit with its reason, the day it happens, not from memory on Friday.
2. Fix the biggest reason first. One cause is often most of the total, as the burnt load was here.
3. Turn the fix into a check on the batch record, with a limit: oil temperature, seal squeeze test.
4. Watch the rate weekly for the reason you fixed, to see the fix held.

Where this stops working

Scrap is the easiest number to lose. The burnt load goes straight into the bin, the torn bags are never counted, and the dropped case is written off as "a bad week". The stock then shows bags that don't exist, the cost per bag looks better than it is, and the scrap rate is whatever someone guesses.

How TaroStack does it

TaroStack records what each batch made good and every loss with its reason, so the scrap rate is counted, not guessed.

Good output on every batch. Recording a batch takes ingredients and packaging off stock from the lots used, at what you paid for each, and divides by what came out good, so scrap during the batch shows in its yield and its cost per unit. Recipe truth shows, from your own batch history, whether a recipe yields what it claims and which ingredient's quantity is wrong.

Every loss, with its reason. Product damaged, spoiled or lost after it's made comes off stock as an adjustment with its reason, so waste is reported by reason, item and site, and every movement is a permanent record of who, when and why.

And the checks that prevent it. Batch record forms carry your own fields with limits, like oil temperature or a seal test, so a reading outside its limit is flagged while there's still time to save the load.

Batches, true cost, recipe truth, adjustments with reasons and batch records with limits are on every plan, from $49 a month. The first 30 days are free, with no credit card. If you'd like a hand setting up, ask, and we'll do it with you.

Questions people also ask

How do you calculate scrap rate?

Divide the units scrapped by the units started, and multiply by 100. A week that starts 2,400 bags and scraps 58 has a scrap rate of 2.4%.

What is the difference between scrap and rework?

Scrap can't be sold or fixed and is thrown away. Rework is product that's fixed, like an underweight bag topped up, and then sold. Count them separately.

What is a good scrap rate?

Your own, falling. A scrap rate depends on the product, the process and how scrap is counted, so the useful comparison is this week against last, by reason.

How does scrap affect cost per unit?

The good units pay for the scrapped ones: cost per good unit is the total cost of the run divided by the good units. In the example, 58 scrapped bags in 2,400 add about 5 cents to each good bag.