Feed mill: one tonne is not another

“We sell a tonne of compound feed. For twelve years I thought a tonne was a tonne”

Valentyn Kushnir runs a compound feed mill in Mykolaiv region. The plant is small: 740 tonnes a month, three main formulations, twenty-two people.

He took it over from his father. The mill has stood on the same site for thirty years. One line, one shift, a raw-material store for a week of work.

Customers know him by his prices. Pig feed at 19 800 ₴ a tonne, poultry at 20 400 ₴, cattle feed at 16 200 ₴. Those three numbers held for almost a year.

Last spring soybean meal rose within a month. Valentyn learned it from a supplier invoice, not from his own spreadsheet. By then April was already ending.

What a feed mill sells

The mill sells tonnes. The customer orders tonnes, the manager invoices tonnes, the store ships tonnes. All reporting is built around that unit.

But the tonnes differ inside. Every formulation has its own components in its own proportions. Those proportions are approved by the technologist, not by sales.

One tonne can be mostly corn. Another is half expensive meal. The price list meanwhile carries one figure for the whole year.

The review happens once a season, in spring. Between reviews the price lives its own life, and the raw materials live theirs.

You do not sell tonnes. You sell a formulation that weighs a tonne.

While input prices hold still, the difference between formulations stays invisible. The mill looks like one production with three price tags.

In fact these are three different economies on one line. Each has its own cost structure and its own sensitivity to the market.

Sensitivity is the key word. One formulation leans on soy, the second on sunflower, the third on wheat.

Three formulations, one table

Valentyn put the three formulations into one table for the first time, with composition and not only prices. The proportions had always been known, but they sat apart in technology cards.

The cards were kept by the technologist, the price list by the commercial director. Those two documents met once a year, at a meeting.

The difference jumps out at once. Two formulations hold soy, the third holds none at all. The technologist explained that by the needs of the animals, and money never came up.

In its place the cattle formulation carries sunflower meal — 40% of the mass. That is the largest share of a single component in the whole range.

The reason is purely technological. Ruminants take up sunflower protein differently from pigs or poultry.

A technologist made that call twenty years ago. Nobody looked for financial sense in it then.

Twenty years later the same call saved the mill a quarter. The technologist never suspected the credit.

Input prices

Next Valentyn wrote out the price of every item. All of them come from delivery notes of the same week, with no quarterly averages and no forecasts.

• Corn — 9 200 ₴ a tonne
• Wheat — 8 600 ₴
• Soybean meal — 21 500 ₴
• Sunflower meal — 11 400 ₴
• Premix — 86 000 ₴
• Feed-grade chalk — 2 400 ₴

The gap between the cheapest and the dearest item is thirty-six times. That gap is what makes formulations unlike each other in cost, although they weigh the same tonne.

“Premix costs as much as nine tonnes of corn. But it is five per cent of the formulation, and nobody thinks about it”

Share and price pull in opposite directions. A cheap component with a large share weighs more in cost than a dear one with a small share.

Corn in the pig formulation gives 5 060 ₴ a tonne, premix gives 4 300 ₴. And premix is nine times dearer.

Cost per tonne

Multiplying proportions by prices gave three different numbers. A five-minute job the mill never did on a regular basis.

It was done only when a new formulation was launched. After that, cost lived on as one average number for the whole plant.

Raw material for the pig formulation costs 15 452 ₴ a tonne. For poultry it is 16 271 ₴. For cattle it is 12 738 ₴. Three formulations, three different tonnes.

Valentyn checked the calculation twice. The numbers did not match what he had carried in his head for years.

In his head all three formulations cost about the same. The difference between them felt like a question of premix, not of structure.

Premix is indeed the dearest per tonne. But its share is small, and it moves the total less than corn does.

The gap between the dearest and the cheapest formulation is 3 533 ₴ a tonne.

That is more than a fifth of the selling price. No markup can stay the same across such a gap.

“Three and a half thousand of difference on one tonne. And the price list carried one markup for all three”

Wheat at 15% and soy at 15% are different rows

In the pig formulation wheat is 15%. After the rewrite soy will stand there at 15% too.

The same share, different components, a different price per tonne. They must not be confused.

Processing is the same

On top of raw material comes processing: grinding, pelleting, energy and labour. That is everything that happens to the components after weighing.

Valentyn counts it at 1 450 ₴ a tonne for every formulation. The line works the same with soy and without it, and pelleting time does not move.

That is convenient for accounting and dangerous for intuition. One identical row suggests the tonnes are identical too. Here is where the picture of the business parts company with its economics.

Accounting works with a single processing rate as well. In the monthly report it stands as one line against the whole volume.

Nobody asked to split that line by formulation. And the split would bring no benefit — the rate really is the same.

Processing is the same. Cost per tonne is not.

Margin by formulation

The ranking looked obvious. The pig formulation led with a margin of 2 898 ₴, the cattle formulation came last with 2 012 ₴.

Valentyn sold most of the cattle feed: 320 tonnes a month against 240 of pig feed. Poultry moved least — 180 tonnes.

The mix had formed historically, around the customers of the district. There are more dairy farms nearby than pig complexes.

It irritated him every month. The largest volume brought the smallest margin per tonne. It looked like a structural error in the portfolio.

Valentyn raised the price of the cattle formulation twice. Both times part of the customers went to the neighbouring mill.

The cattle feed market is more sensitive to price. Farms count the cost of a litre of milk every day.

Feed is the largest cost item there. Any increase shows up in their own economics at once.

Pig complexes count no less carefully, but their cycle is longer. The reaction to a price change arrives weeks later.

“For two years I worked on moving customers from cattle feed to pig feed. Good thing I never got round to it”

The month in numbers

Together the mill makes 740 tonnes a month. Revenue comes to 13 608 000 ₴ with the line fully loaded.

The cost of all three formulations is 11 786 420 ₴. Gross profit works out at 1 821 580 ₴ a month.

Fixed costs at the mill are 980 000 ₴ a month. Net profit comes to 841 580 ₴ — that is 6.2% of revenue, a normal figure for compound feed.

Volume and margin rank differently

Cattle feed gives the largest volume and the smallest margin per tonne. Pig feed does the opposite.

While prices are stable, those two rankings do not argue. The mill earns on volume and on margin, just in different formulations.

The argument starts when one component gets dearer. Then volume and margin part ways for good.

Drying shrinkage nobody sees

Before the formulations comes grain intake. Grain arrives wetter than it goes into production, and the dryer takes the difference away.

Out of 1 000 tonnes of grain received, 966 tonnes reach the formulation. Drying shrinkage is 34 tonnes, or 3.4%.

The figure is stable year after year. It depends on the weather at harvest, not on the work of the mill.

In money, at the corn price, that is 312 800 ₴ for every thousand tonnes received. The sum matches a third of monthly fixed costs.

The loss is normed and unavoidable. The problem is not the loss itself, but that it rarely enters the cost per tonne.

The mill buys wet grain at the price of grain. The moisture leaves in the dryer, and the money for it has already been paid.

In most calculations that row is simply absent. Cost is counted on the mass that came out, while payment covers the mass that came in.

A normed loss is still a cost. Being normed does not make it free.

34 tonnes are not spoilage

The grain did not spoil and was not stolen. It gave up moisture that the mill had paid for as grain.

Soy gets dearer

In April soybean meal moved from 21 500 ₴ to 26 800 ₴ a tonne. The supplier gave a week of warning, and the mill held no fixed-price contract.

Plus 5 300 ₴ on a tonne, almost a quarter of the price. The rest of the items stayed put, and that made the consequences very uneven.

The sales managers came to Valentyn with one question. How far should the price list go up to hold the margin?

“They asked for a percentage for the whole price list. But the answer was different for every formulation”

The question looked technical. In fact it was the most important one of that quarter.

The same percentage across the whole list would mean two mistakes at once. A shortfall where prices rose, and an extra increase where they did not.

What happened to the formulations

The loss turned out directly proportional to the soy share. A quarter of soy in the poultry formulation cost the most — 1 325 ₴ out of every tonne.

The cattle formulation did not change at all. It holds not a kilogram of soy, so the increase simply does not reach it.

2 012 ₴ stands in the table twice

This is not an error and not a repeat. The margin of the cattle formulation before the increase and after it is the same number.

That lack of change is the main fact of this table.

Owners are used to hunting for changes in a table. Here the row that matters is the one where nothing moved.

The soy-free formulation won

The ranking turned over in one month. No decision had to be taken for that.

The turn happened in somebody else's price list, a few hundred kilometres from the mill. Valentyn was the last to hear of it.

The formulation that had irritated him for two years with the smallest margin became the most profitable in the range. The same 2 012 ₴ a tonne was now the largest number in the table.

It did not get better. All the others simply got worse, because they leaned on a component whose price the mill does not control.

Valentyn put it to himself differently. The soy-free formulation turned out not the most profitable, but the most resilient.

Profitability and resilience are different properties. In a calm year you cannot tell them apart.

Resilience shows up not in a calm year, but when inputs jump.

What the increase cost

Gross profit fell from 1 821 580 ₴ to 1 303 240 ₴ on the same production volume.

The difference is 518 340 ₴ a month. It showed in none of the reports Valentyn read every week.

Fixed costs stayed where they were — 980 000 ₴. Net profit came to 323 240 ₴.

That is a fall of 61.6% in one month. Not a single line in the shipping report moved.

Revenue did not change either. No customer left, not a tonne stayed in the store.

That is what makes such losses invisible. All the usual indicators of the mill looked normal.

Shipments on schedule, the store empty, no complaints. No profit either.

None of the classic operating indicators can show this. They measure movement, not money.

“Sales were the same. Two thirds of the profit was gone. None of it showed in the shipping report”

First move: count the formulation

Valentyn set a rule for the planning and economics department. It holds whether the month is calm or not.

Cost is recalculated on delivery-note prices every week, not once a quarter. A quarter is far too long to catch a jump in one item.

A week is no ideal either. But in a week the mill ships 185 tonnes, and in a quarter more than two thousand.

The job takes half an hour. The table is already built, and only six prices from the delivery notes change.

• Six input prices from the delivery notes of the week
• Three formulations with fixed proportions
• Cost and margin of every formulation at the output

The department resisted at first. A weekly recalculation looked like extra reporting.

The resistance went after the very first month. The table showed a move in premix prices that nobody expected.

Before, those numbers appeared after the fact, in the monthly report. Now they stand before the price decision, not after it.

Second move: look at structure

The second decision covered negotiations with customers. It proved harder than the first, because it touched the managers.

Valentyn stopped lifting the price list by one percentage. Every formulation got its own conversation and its own figure.

For the managers that meant more work. Instead of one letter to all customers they prepared three different justifications.

The cattle feed customer got no increase at all. His formulation had not grown dearer, and he would have checked that within a week.

Pig and poultry customers were lifted differently. The pig formulation got less, the poultry one more.

Neither of them left. Both watched the soy market and knew where the figure came from.

One percentage for the whole list admits you never counted cost.

The customer can count too

Large farms know soy prices no worse than the mill does. An increase where nothing grew dearer is spotted at once.

Trust is dearer than a one-off shortfall. A customer who catches a supplier lifting prices without grounds will check every invoice after that.

Third move: rewrite the formulation

The technologist offered to replace part of the soy with sunflower meal. The proposal had lain with him for a long time, but before the increase it made no sense.

At the old soy price the swap would make the formulation worse with no gain in money. The increase changed that balance.

Soy in the pig formulation went from 22% to 15%. Sunflower meal added 9%, feed-grade chalk dropped to 1%. The rest of the components stayed as they were.

• Raw material: was 15 452 ₴, after the increase 16 618 ₴, in the rewritten one 15 720 ₴
• Cost: was 16 902 ₴, after that 18 068 ₴, in the rewritten one 17 170 ₴
• Margin: was 2 898 ₴, after that 1 732 ₴, in the rewritten one 2 630 ₴

The rewrite won back 898 ₴ of the 1 166 ₴ lost. A little more than three quarters.

The remaining 268 ₴ a tonne had to be accepted. They could be recovered through the price list only.

Valentyn decided against that. Two increases in one quarter would cost more than the shortfall.

Customers forgive one justified increase. A second one in a row they read as a reason to look at competitors.

Gross profit rose to 1 518 760 ₴, net profit to 538 760 ₴. The mill came back to a working mode, but not to the old one.

That is still below the original 841 580 ₴. An increase is never won back in full — a formulation cannot make soy cheaper.

A formulation is not only money

Sunflower meal is cheaper, but its protein is different — in composition and in digestibility.

If weight gain at the customer falls, he will count it faster than the mill. And he will go to someone who left the formulation alone.

Valentyn agreed a trial batch with two farms. The rewritten formulation went into production only after they confirmed the figures.

The rest of the customers had the old formulation at the new price for two months. That cost money, but it kept the relationships.

“I have no right to save on somebody else's weight gain. That is a saving for one season and a customer lost for good”

What to do on Monday

Valentyn boiled it down to three actions. None of them touches production.

• Count the cost of every formulation at the prices of this week
• Write out what share the dearest item takes in each formulation
• Before lifting the price list, check whether that formulation grew dearer

No action calls for a new line or new equipment. All three are done in the same table.

The table sits in the planning and economics department and takes one sheet. Updating it is routine, not a project.

All three change what the manager brings to the customer.

The first two are done once and then repeated every week. The third is a question asked before every change to the price list.

The third action proved the hardest. It calls for a stop at the moment when the decision looks obvious.

Obvious decisions in compound feed cost the most. They are the ones taken fast and without the table.

A price list blind to cost protects somebody else's profit, not yours.

The money does not vanish — you simply do not see it

At the mill nothing visible changed that April. The change sat in one figure on somebody else's invoice.

The same line, the same 740 tonnes, the same customers and the same shipping notes.

One figure in somebody else's price list moved. Everything else stayed in place, and that is why the loss was so easy to miss.

Two thirds of the profit were gone. It could be seen in one place only — in the cost of the formulation, counted at fresh prices.

The mill did not get worse and did not lose a single customer. It spent two months selling at old prices what had grown dearer.

A tonne is a unit of weight. Profit lives in the formulation.

How to see this at your own mill

Everything above rests on one row: not the tonne at a single price, but each recipe on its own — with the share of every component and the price from this week's delivery note. As long as cost is averaged across the plant, you see neither that recipes differ by 3 533 ₴ a tonne, nor that a rise in one input hits them unequally. When the same data is collected automatically, it becomes visible what each recipe earns today — which is exactly why owners of small and medium businesses use Finmap: to see the financial picture rather than assemble it at the end of the month.

Read also

• Poultry farm: three identical houses, different results
• Dairy farm: a cow is milked 305 days and eats 365

Topic foundation: Management Accounting Explained: What It Is and Why Owners Need It