Making a kilogram of cheese leaves about nine kilograms of whey, a liquid holding the milk’s lactose and a fifth of its protein. India imports most of the whey protein its gyms and nutrition brands use, at prices that roughly tripled between 2024 and mid-2026. Yet the government’s model project report for a small cheese plant, written to guide subsidised loans under the PM Formalisation of Micro Food Processing Enterprises scheme (PMFME), does not account for whey at all. The fix is not a protein dryer in every dairy: a small plant’s whey holds only a few tonnes of protein a year, far too little to pay for one. It is to make whey a costed part of every small cheese project, pool the rest in shared whey hubs, and let the large plants that already dry whey take in their neighbours’.

How much to trust it. The plant figures come from the project report itself and the whey composition from a US Department of Agriculture technical report; where this piece calculates, it says so. The Kenyan returns are from a 2025 preprint that has not yet been peer reviewed. India-specific figures on imports, prices and companies are from press reports and could not be checked against primary sources; several conflict, and the piece says where. This follows this blog’s earlier look at cheese and value-added dairy.

The model plant counts the cheese and forgets the whey

The project report, prepared by the National Institute of Food Technology Entrepreneurship and Management (NIFTEM) for the food processing ministry, sizes a cheese unit at 400 kg a day: Rs 33.61 lakh in all, Rs 23.05 lakh of it machinery, with output rising from 40% to 60% of capacity over five years and a net margin of 1.2% to 4.1% of sales. It costs milk, power, wages, packaging and rent. Whey appears only as a step in the process chart, “draining of whey”.

Using the USDA report’s figures, about 9 kg of whey per kg of cheese, of which 0.8% is protein and 4.9% lactose, the plant would produce about 390 tonnes of whey in its first year and about 580 tonnes in its fifth, roughly 1,300 to 1,900 kg a day. That is nine times the weight of the product the plan is built around, and a unit that pours it away would struggle to meet the dairy effluent limit of 100 mg/L of biochemical oxygen demand. The USDA report notes that ordinary dairy wastewater already runs at 2,000 to 3,000 mg/L.

Why a dryer in every dairy does not add up

The same whey holds only 3 to 5 tonnes of protein a year at the model plant’s output, and about 8 tonnes at full capacity. Turned into WPC80, an 80%-protein powder, that is roughly 3 to 8 tonnes, worth about Rs 0.8 crore to Rs 2.5 crore a year at mid-2026 import prices, by this blog’s arithmetic. The only published cost found for an ultrafiltration, diafiltration and drying line to make WPC80 is a European one, EUR 7-11 million. No Indian figure for the cost or minimum scale of such a plant was found.

The economics change with volume. Milky Mist, a Tamil Nadu dairy, reports about 1 million litres of cheese whey a day and is planning a whey protein plant. At about 8.5 kg of WPC80 per 1,000 litres, a stream that size could yield on the order of 2,500 tonnes a year. A few hundred units the size of the model plant together produce that much whey, which is the case for pooling it.

The market is there. The USDA’s market report priced WPC80 at US$12.50 to 13.00 a pound in mid-July 2026, and Indian importers and brands report landed prices rising from Rs 700-800 a kg in 2024 to Rs 2,300-3,000 in mid-2026. Estimates of India’s protein-supplement market range from about US$0.9 billion to US$1.2 billion, and 80-90% of supplement-grade whey is reported to be imported, mainly from the United States, New Zealand, the European Union and Australia. Import volumes are uncertain: one report puts them above 45,000 tonnes in 2022-23, another estimate at about 23,000 tonnes in 2025. India’s trade deals have so far left dairy alone, as this blog found for the New Zealand agreement, which excludes whey.

What small units can do: the Kenyan evidence

A 2025 cost-benefit study of four Kenyan dairies found every small whey product profitable: ricotta from whey alone, ricotta from whey and milk, a whey drinking yoghurt made half from whey, and a whey and fruit smoothie. Equipment cost from about US$5,000 to about US$100,000, and counting avoided effluent treatment, energy recovery, emissions and jobs added 12-18% to the social return. India has its own recipes: the ICAR-National Dairy Research Institute lists whey drinks flavoured with jaljeera, tamarind and mango as ready to license.

The limit is volume. The four Kenyan processors together used about 145 tonnes of whey a year; one Indian plant of the model size makes 390 to 970 tonnes. Drinks and whey cheese can take part of a small unit’s whey, but not all of it.

A plan by size

Small units should treat whey as part of the cheese project: drinks and whey cheese first, food use and contracted cattle feed next, biogas for what is left, and never untreated discharge. Where a hub exists, they should chill and concentrate whey and sell it, paid by the protein and solids delivered, the way milk is paid by fat and solids-not-fat.

Shared whey hubs, run by milk unions or producer companies where cheese and paneer units cluster, would do the drying. Reverse osmosis at each unit removes water, which is about 93% of whey, before it is trucked; the hub separates protein from lactose and dries both. The credit already exists: the Animal Husbandry Infrastructure Development Fund lends up to 90% of project cost with a 3% interest subsidy, though, as this blog found, only 39% of what it has sanctioned has been disbursed. The lactose-rich permeate left after protein is removed can go to lactose, to biogas, as in the biogas incentive stack, or, as Amul is trialling, to ethanol.

Large plants are already moving. Amul began making WPC80 at its cheese plant in August 2025; Parag Milk Foods built its cheese and whey plant at Manchar for Rs 110 crore and runs a 40 tonne-a-day lactose plant; Bhole Baba Dairy reports 150 tonnes a month of whey protein. They can also buy concentrated whey from small units nearby, turning a neighbour’s disposal problem into their feedstock.

The paneer problem

Most of India’s whey probably does not come from cheese at all. Paneer and chhena, made by curdling milk with acid and heat, leave an acid whey, and one 2025 review estimates India’s total whey at 3.3 to 5 million tonnes a year, derived from paneer output. Some 90-94% of paneer is made by unorganised producers, so most of that whey is scattered across small units and not collected. Acid whey carries more minerals and lactic acid than cheese whey and is harder to turn into protein powder; for paneer makers, drinks, feed and biogas are the realistic outlets, and paneer clusters are the obvious first sites for hubs. No source measures how much of this whey is used.

How big the market is, and who builds these plants

Demand. Forecasts for India’s whey protein market are all paid market-research estimates, and they disagree: bases of US$100-190 million in 2024-25 and growth of 3-6% a year to roughly US$125-220 million by 2030-31. They were mostly made before the 2026 price rise and give values, not tonnes, so they say little about how much protein India will need. The wider categories that use whey are forecast to grow faster.

Prices are the clearer signal. Rabobank, as reported in the trade press, finds milk supply in the main exporting regions slowing while demand for protein grows, driven by high-protein foods and by people on GLP-1 weight-loss drugs told to eat more protein; strong cheese output last winter did not ease whey prices. New capacity abroad arrives late: Glanbia’s expansion in 2027 and Agropur’s by 2029. The USDA did see US demand soften in July 2026, so tight supply through 2026 and into 2027 is well supported, but nothing beyond that can be forecast with confidence. In India, announced supply is thin: Parag processes about 10 lakh litres of whey a day, Milky Mist plans a protein plant for about 1 million litres a day, and Lactose India makes about 10,000 tonnes of lactose a year with plans to double. Nobody publishes India’s total whey protein capacity, so the gap between demand and domestic supply cannot be measured.

Builders. India has the firms to build all three tiers, though whey-specific Indian references are scarce.

A likely team for a shared whey hub would be NDDB Engineering or IDMC as project manager and contractor, an Indian drying supplier and a membrane integrator; large dairies have tended to buy complete lines from global suppliers. No Indian price for a whey plant was published: the nearest benchmarks are about US$21 million for a 1,000 kg-an-hour WPC80 line in a Romanian study and the 15-18 months Milky Mist expects to take to build its plant. Firms are listed only where this blog found evidence of dairy or whey work; this is not an endorsement, and buyers should ask each for Indian whey references.

What is not known

1. Volumes. India publishes no tonnage for cheese, paneer or chhena, and no estimate of whey used or discharged.
2. Duty. The customs duty on whey (HS 0404.10) was not found, and sources disagree on the duty on whey protein concentrate (HS 3502.20).
3. Plant cost. No Indian capital cost or minimum economic scale for a whey protein line was found.
4. Prices. Whey protein prices roughly tripled in two years; any investment case should be tested at 2024 prices too.

What would fix this

• NIFTEM should add a whey section to its model cheese and paneer project reports, with the mass balance, an outlet, its equipment and its revenue or disposal cost, and correct the milk price and chiller size in the cheese report.
• The food processing ministry should let PMFME subsidy cover whey equipment bought as part of a cheese or paneer project.
• The Department of Animal Husbandry and Dairying should name whey hubs as eligible under AHIDF and publish annual cheese, paneer and chhena output with an estimate of whey generated and used.
• The food processing ministry should add whey protein and lactose to the products eligible under the PLI scheme, alongside mozzarella.
• State pollution control boards should enforce the dairy effluent limit with a grace period tied to a whey plan, so that disposal stops being free.

These are proposals from this blog, not recommendations any agency has adopted.

Sources and caveats

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.