---
title: Yield, wastage and by-product formulas every food factory should track
canonical: https://naffo.tech/blog/yield-wastage-formulas-food-manufacturing
question: What are the formulas for yield, wastage and by-product recovery in food manufacturing?
published: 2026-03-04
updated: 2026-08-08
author: naffo.tech manufacturing team (Plant systems and costing)
reviewed_by: naffo.tech implementation desk (Dairy and food plant rollouts)
publisher: naffo.tech — https://naffo.tech
category: Costing & metrics
tags: yield, wastage, formulas, dairy, costing, by-product
reading_time_minutes: 10
license: Free to quote with attribution to naffo.tech (https://naffo.tech/blog/yield-wastage-formulas-food-manufacturing)
---

# Yield, wastage and by-product formulas every food factory should track

**Question:** What are the formulas for yield, wastage and by-product recovery in food manufacturing?

**Answer:** Yield % is good sellable output ÷ total input × 100. Wastage % is actual wastage ÷ total input × 100, excluding recoverable by-product. By-product recovery % is recovered by-product ÷ theoretical by-product × 100. Wastage cost is wasted quantity × material cost plus the value added up to the loss point. Every one of these needs a stated basis (kg per litre, or kg per kg) and a fixed rule on whether rework and by-product count as output.

Naffo is independently developed. Tally and TallyPrime are trademarks of Tally Solutions Pvt. Ltd. Other company and product names belong to their respective owners and are used to identify products or describe compatibility. References do not imply affiliation, endorsement or certification. Integration availability depends on the product version, plan, permissions and configuration; confirm your required workflow before purchase.

## Key takeaways

- A yield number without a stated basis is meaningless. "18% paneer yield" only means something as kg of paneer per litre of milk at a stated fat percentage.
- Exclude recoverable by-product from wastage and include it in a separate recovery metric — otherwise a well-run paneer plant looks like it wastes 76% of its milk.
- Cost loss at the value it had when it was lost, not at raw-material rate. Loss after packing carries material, energy, labour and film.
- Theoretical yield comes from the recipe; actual yield comes from the batch. The gap, expressed in rupees, is your improvement budget.
- Fat- and solids-corrected yield is the only fair comparison in dairy, because input quality moves daily and volumetric yield moves with it.
- Give-away (systematic overfill on a filler) is a real loss category with its own formula and usually its own uninvestigated cost.

## Key figures

- **Always** — Fat-corrected basis is required for fair yield comparison in dairy (Basis: Volumetric yield moves with incoming fat and SNF; comparing raw litre-based yields across days compares milk quality, not plant performance.)

## Fix the definitions before you fix the numbers

In most factories, production, QC and accounts each report a different loss figure for the same batch — and each is arithmetically correct, because they are using different definitions. Three questions settle it, and they must be answered once, in writing, for the whole plant:

1. **Is rework output or loss?** Rework is neither: it is a temporary state. Count it separately at batch close and account for it in the batch it is eventually consumed in. Counting rework as good output inflates yield; counting it as loss double-counts when it is later sold.
2. **Is by-product output?** Yes, but not in the yield of the main product. Whey is not paneer. Give it its own item code, its own rate and its own recovery metric.
3. **On what basis is yield expressed?** Output kg per input litre, output kg per input kg, or fat-corrected equivalents. State it in the report header, permanently.

> **WARNING**
>
> The single most common reporting error in dairy is treating whey as wastage. A paneer plant converting 1,000 L of milk into 180 kg of paneer and 760 L of whey does not waste 76% of its input — it recovers most of it as a by-product with its own market. Getting this wrong makes every wastage report unusable.

## The core formulas

**1. Yield %**

```
Yield % = Good sellable output ÷ Total input × 100
```
Good sellable output excludes rework, rejection and by-product. Always publish the basis alongside the number.

**2. Process loss %**

```
Process loss % = (Input − Good output − By-product − Rework − Rejected) ÷ Input × 100
```
This is what genuinely disappeared in the process — evaporation, adherence to vessels, drain, trim. It should be stable batch to batch; instability means process variation, not material variation.

**3. Wastage %**

```
Wastage % = Actual wastage quantity ÷ Total input quantity × 100
```
Actual wastage = spillage + contamination + expiry + unrecoverable rejection. It excludes recoverable by-product and excludes rework that will be consumed.

**4. Wastage cost**

```
Wastage cost = Wasted quantity × (Material cost + Value added up to loss point)
```
Value added includes energy, labour, packaging and overhead already absorbed. A kilo lost at intake and a kilo lost after sealing are not the same rupee.

**5. By-product recovery %**

```
By-product recovery % = Recovered by-product ÷ Theoretical by-product × 100
```
Theoretical by-product comes from the recipe. Recovery below about 95% usually means drain loss at transfer points.

**6. Yield variance vs standard**

```
Yield variance = (Actual yield % − Standard yield %) ÷ Standard yield % × 100
```
Report this signed. A positive variance is not automatically good news — it can indicate under-dosing, moisture retention or a weight-spec breach that QC will catch later.

**7. Rework rate**

```
Rework rate % = Rework quantity ÷ Total output (good + rework + rejected) × 100
```
Rising rework with flat rejection means the process is drifting but QC is catching it. That is a warning, not a success.

**8. Rejection rate**

```
Rejection rate % = Rejected quantity ÷ Total output × 100
```
Break down by reason code. An aggregate rejection rate tells you the size of the problem and nothing about the cause.

**9. Give-away (overfill) %**

```
Give-away % = (Average actual fill weight − Declared weight) ÷ Declared weight × 100
```
Legal metrology requires you to stay above the declared weight, so some overfill is deliberate. Systematic overfill beyond the required tolerance is free product. On a 200 g pack running 2 g heavy, that is 1% of everything you ship.

**10. Packaging loss %**

```
Packaging loss % = Packaging material damaged or scrapped ÷ Packaging material issued × 100
```
Count units for pouches, bottles, caps, labels and cartons; count metres for film. Changeovers are the biggest single driver.

**11. Expiry write-off %**

```
Expiry write-off % = Value of expired stock ÷ Value of finished goods produced in the period × 100
```
Track by SKU, not in aggregate. Expiry is almost always concentrated in a few slow movers that the plan keeps producing on autopilot.

**12. Overall material efficiency**

```
Material efficiency % = (Good output value + By-product value) ÷ Total input value × 100
```
The one number for the owner. It is denominated in rupees, so it cannot be gamed by unit tricks, and it moves only when something real improves.

## Worked example — paneer

> **Paneer batch, 1,000 L milk**
>
> Input: **1,000 L** buffalo milk at **6.0% fat**, cost **₹52/L** → input value **₹52,000**. Standard yield: **18.5 kg per 100 L**.
>
> Actual output: good paneer **181 kg**; whey tanked **755 L** (theoretical 780 L); rework **4 kg**; QC rejection **2 kg**; measured spillage **5 L** milk equivalent.

_Paneer batch — metric by metric_
| Metric | Calculation | Result |
| --- | --- | --- |
| Yield % | 181 kg ÷ 1,000 L | **18.1 kg / 100 L** |
| Yield variance | (18.1 − 18.5) ÷ 18.5 × 100 | **−2.2%** vs standard |
| By-product recovery % | 755 ÷ 780 × 100 | **96.8%** |
| Rework rate | 4 ÷ (181 + 4 + 2) × 100 | **2.1%** |
| Rejection rate | 2 ÷ 187 × 100 | **1.1%** |
| Actual wastage | 5 L spillage + 2 kg rejection (≈ 11 L milk equivalent) | **≈ 16 L equivalent** |
| Wastage % | 16 ÷ 1,000 × 100 | **1.6%** |
| Wastage cost | 16 L × ₹52 (plus conversion on the rejected paneer) | **≈ ₹950 per batch** |
| Yield shortfall cost | 4 kg short × ₹310/kg realisable | **≈ ₹1,240 per batch** |

The instructive part is the last two rows. The visible wastage costs about ₹950; the invisible yield shortfall costs about ₹1,240 — and it appears in no wastage register anywhere. **Across 25 batches a month that is roughly ₹31,000 of product that was never lost, spilled or rejected. It simply never formed.** Yield variance, not wastage, is usually the bigger number.

## Worked example — ghee

> **Ghee from cream**
>
> Input: **500 kg** cream at **60% fat** = 300 kg fat. Ghee is roughly **99.5% fat**, so theoretical ghee ≈ **301 kg** at full fat recovery; a realistic standard allows a **1.5–2%** fat loss to residue and vessel adherence.
>
> Actual: ghee **293 kg**; ghee residue collected **9 kg** (sellable by-product); balance is moisture driven off plus unrecovered fat.

- **Fat-basis yield** = 293 × 0.995 ÷ 300 × 100 = **97.2% fat recovery**. This is the only honest ghee yield metric, because cream fat varies daily.
- **By-product**: 9 kg of residue at a realisable rate is a recovery line, not wastage. Plants that drain residue lose a small but perfectly recoverable margin every batch.
- **Do not** report ghee yield as 293 ÷ 500 = 58.6%. That number falls whenever cream fat falls, and the plant gets blamed for the supplier's milk.

## Worked example — curd and bakery

_Two more common cases, and the metric that actually matters in each_
| Product | Input | Output | The metric that matters |
| --- | --- | --- | --- |
| Set curd, 200 g cups | 1,000 L standardised milk | 4,850 cups filled, 60 cups rejected for seal failure | **Fill efficiency** = 4,850 × 0.2 ÷ 1,000 × 100 = 97% of milk into saleable cups; plus **packaging loss** on the 60 seal failures, which is a machine problem, not a milk problem |
| Bread, 400 g loaves | 100 kg flour + water, fat, yeast per recipe | 168 loaves, 5 kg dough trim, 3 loaves under-weight | **Baker's yield** (loaves per 100 kg flour) plus **dough trim recovery %** — trim reworked into the next batch is recovery, trim binned is wastage |
| Frozen snack, 500 g packs | 250 kg prepared mix | 470 packs, 6 kg broken product, 2.1 kg average overfill | **Give-away %** = 2.1 ÷ 235 × 100 ≈ 0.9% — invisible in every stock report and worth more than the broken product |

## Standard vs actual: where the improvement budget comes from

Theoretical (standard) yield comes from the recipe and the science. Actual yield comes from the batch. The disciplined version of continuous improvement is simply: keep a standard, measure the gap in rupees, attribute the gap, close the biggest one.

**Annualised value of a one-point yield gain**

```
Annual gain = (Yield gain % ÷ 100) × Annual input quantity × Realisable value per output unit
```
Worth calculating before any capex conversation. On 3,00,000 L of milk a year at ₹310/kg paneer, a single percentage point of yield is roughly ₹9.3 lakh — which reframes what a ₹2 lakh press upgrade is worth.

The attribution step is where a system beats a spreadsheet. Yield has to be sliceable by product, recipe version, shift, operator, machine and incoming lot; otherwise you know the gap exists and not where it lives.

## Reporting rules that keep the numbers trustworthy

- **One basis per product, stated in the report header.** Never mix kg-per-litre and kg-per-kg in the same table.
- **Fat- or solids-corrected in dairy.** Compare like with like or you are measuring your milk supplier, not your plant.
- **Rupees next to every percentage.** Percentages set priorities badly; a 0.4% packaging loss can outrank a 3% whey loss in cost.
- **Batch-level granularity retained.** Averages are for trends; batches are for causes.
- **Reason codes mandatory, free text optional.** Anything you cannot group, you cannot fix.
- **Recipe version stamped on every batch.** Otherwise a yield step-change cannot be attributed to the recipe change that caused it.

Next: [the batch-wise control system these formulas plug into](/blog/reduce-wastage-food-factory), and [the dashboard an owner should read in five minutes](/blog/owner-dashboard-manufacturing-business).

## Frequently asked questions

### What is the formula for yield percentage in food manufacturing?

Yield % = good sellable output ÷ total input × 100, where good sellable output excludes rework, rejection and by-product. The basis must be stated — for example kg of paneer per 100 litres of milk. In dairy, correct for incoming fat or total solids, otherwise day-to-day yield movement reflects milk quality rather than plant performance.

### How do you calculate paneer yield from milk?

Divide good paneer output by milk input and express it per 100 litres: 181 kg from 1,000 L is 18.1 kg per 100 L. Compare against a standard yield for the same milk fat percentage, and report the variance. Whey is recorded separately as a by-product with its own recovery percentage, never as wastage.

### What is the difference between process loss and wastage?

Process loss is material that inherently disappears in the process — evaporation, vessel adherence, trim — and should be stable batch to batch. Wastage is avoidable loss: spillage, contamination, expiry and unrecoverable rejection. Mixing them makes the number impossible to act on, because process loss is reduced by engineering and wastage is reduced by discipline.

### What is give-away in food packaging?

Give-away is the systematic overfill above declared weight on a filling line. Legal metrology requires you to stay above the declared weight, so a small margin is intentional, but overfill beyond that tolerance is free product. Give-away % = (average actual fill − declared weight) ÷ declared weight × 100. Two grams on a 200 g pack is 1% of everything you ship, and it appears in no stock report.

### How do you cost wastage correctly?

Cost it at the value the material had at the point it was lost: material cost plus energy, labour, packaging and overhead already absorbed. Costing all loss at raw-material rate understates finished-goods loss substantially, which is why plants often under-prioritise packing-line and warehouse losses that are far more expensive per kilo than intake losses.

### Should rework be counted in yield?

No. Rework is a temporary state, not output. Record it separately at batch close and account for it in the batch that eventually consumes it. Counting rework as good output overstates yield; counting it as loss double-counts once it is sold. A rising rework rate with a flat rejection rate is an early warning that the process is drifting.

## References

- [Legal Metrology (Packaged Commodities) Rules — declared quantity and permitted error, Government of India](https://consumeraffairs.nic.in/organisation-and-units/division/legal-metrology)
- [FSSAI — standards for milk and milk products](https://www.fssai.gov.in/)

## Related articles

- https://naffo.tech/blog/reduce-wastage-food-factory
- https://naffo.tech/blog/ideal-erp-workflow-food-manufacturing
- https://naffo.tech/blog/owner-dashboard-manufacturing-business

---

Published by naffo.tech, an all-in-one business management and manufacturing ERP platform for Indian SMEs — GST compliance, invoicing, inventory, batch manufacturing, yield and wastage control, and double-entry accounting in one system. https://naffo.tech

Markdown source: https://naffo.tech/blog/yield-wastage-formulas-food-manufacturing/markdown