First pass yield = units that pass a step the first time ÷ units that went into it. A unit that had to be fixed and a unit that was thrown away both count against it. Rolled throughput yield = the first pass yields of every step, multiplied together: the share of units that went through the whole process without being touched twice. A candle maker pours 96 jars and ends up with 93 good candles, a 96.9% final yield that looks fine. But only 83.8% went from pour to box without a fix, and the fixes took 44 minutes. Below is each formula, the batch step by step, what the gap between the two yields costs, and a free calculator.
The short version
- First pass yield = (units in − units reworked − units scrapped) ÷ units in, for one step.
- Rolled throughput yield = first pass yield of step 1 × step 2 × step 3 …
- Final yield = good units out ÷ units in; it hides the rework.
- Rework costs minutes; scrap costs materials and minutes.
- The step with the lowest first pass yield is where to look first.
- TaroStack records good and rejected units on every batch, and your own fields for the rest. More below.
The formulas
| Yield | Formula | The candle batch |
|---|---|---|
| First pass yield, pouring | (96 − 9 reworked − 2 scrapped) ÷ 96 | 88.5% |
| Rolled throughput yield | 88.5% × 98.9% × 95.7% | 83.8% |
| Final yield | Good out ÷ units in | 93 ÷ 96 = 96.9% |
The difference between the last two is the rework: candles that ended up good, but only after someone fixed them. How the final yield fits into a batch's cost is in batch yield formula.
An example: a batch of 96 candles
Lehua Glow, a made-up soy candle maker on Oʻahu I use in these examples, pours 96 nine-ounce jars and follows them through three steps:
| Step | Units in | Reworked | Scrapped | Passed first time | First pass yield | Passed on |
|---|---|---|---|---|---|---|
| Pouring | 96 | 9 | 2 | 85 | 88.5% | 94 |
| Curing and trimming wicks | 94 | 0 | 1 | 93 | 98.9% | 93 |
| Labeling and packing | 93 | 4 | 0 | 89 | 95.7% | 93 |
At pouring, nine candles came out with sinkholes or wet spots and needed a heat gun or a top-off, and two jars cracked. At curing, one wick had drifted too far off center to fix. At labeling, four labels went on crooked and were redone. Each step passes on what it didn't scrap, so the next step starts with fewer. Multiplied, 88.5% × 98.9% × 95.7% = 83.8%: about 80 of the 96 candles went from pour to box untouched, and 16 were fixed or lost along the way.
What the gap costs
| Cost | |
|---|---|
| Rework: 9 pours at 4 minutes, 4 labels at 2 minutes, 44 minutes at $20 an hour | $14.67 |
| Scrap: 3 jars' wax, fragrance, wick, jar and label at $3.99 | $11.97 |
| The batch | $26.64 |
$26.64 on a batch whose good candles sell for $1,209 wholesale is small; across 50 batches a year it's $1,332, and the rework minutes are minutes not spent pouring. Pouring is the step to work on: its 88.5% is the lowest, and sinkholes often come from pouring too hot or cooling too fast, both of which can be tried and measured on the next batch. How a candle's cost is built is in how to price homemade candles.
First pass yield and final yield
Final yield says how much of what you started with you could sell. First pass yield says how much of it came out right, which is the better measure of whether a process is working. A batch can have a perfect final yield and a poor first pass yield, if everything was fixed in the end. Tracking both shows whether a change to the pour temperature fixed the sinkholes or only moved the work to the heat gun.
Get the calculator
Download the first pass yield calculator. One line for each step: type the units that came in, how many needed rework and how many were scrapped, and the minutes a rework takes; the units passed the first time, each step's first pass yield and the units passed on fill in, with the rolled throughput yield, the final yield, and what the rework and scrap cost at the side. It's an ordinary Excel file with no macros, and it works in Google Sheets, LibreOffice and Numbers.
Where this stops working
One batch on a sheet says how that batch went. Whether pouring's first pass yield is getting better since the thermometer was replaced takes every batch, in order, with the same counts recorded the same way, and the scrapped jars taken off the stock they came from.
How TaroStack does it
TaroStack records good and rejected units on every batch, and your own fields for the rest.
Good and rejected on the batch. Each batch records how much came out good and how much was rejected, so its final yield and its cost on the good units are there for every batch.
Your own fields for rework. A batch record form carries fields of your own, such as the number of candles reworked at pouring and the pour temperature, with limits and a sign-off, so the counts are kept the same way each time. How a batch record works is in what is a batch record.
Scrap with a reason. A cracked jar or a candle that can't be fixed is an adjustment with a reason, damaged, so waste is reported by reason, item and site. From your own batch history, TaroStack also says whether a recipe yields what it claims.
| The job | Notebook | In TaroStack |
|---|---|---|
| Good and rejected candles | Counted, maybe | On every batch |
| Rework at pouring | Not written down | A field on the batch record |
| Cracked jars | Thrown out | Adjustment with a reason |
| Is pouring getting better? | A feeling | Batch history |
Batches, batch records and waste reports 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.