🥛 The Business of Smallholder Dairy

Almost everything else you can grow on a small piece of land pays you once. You plant, you wait, you weed, you harvest, and one day money arrives. Then you wait a year. A dairy cow is different. She pays out daily, or every few days, all the way through her lactation. You milk her in the morning, and the milk is sold that morning at the farm gate or delivered to a cooperative. That steady drip of cash is the honest commercial reason smallholder dairy has spread across East Africa, and it matters far more to a small business than the headline litres figure does.

Think about what daily cash actually buys you. School fees paid in instalments instead of one impossible lump. Salt, soap and paraffin bought without borrowing. Money to buy feed for the cow herself, out of the cow's own earnings, which is the difference between a cow that is fed and one that is not. A business with daily income can survive shocks that kill a business with annual income, because it never has to wait six months for the next money.

Now look honestly at the scale of the thing. A large panel study followed 1,317 cow-owning households in Asembo, Siaya County, western Kenya, from 2013 to 2016, giving 3,682 household-round observations. The average herd size in that sample was 1.9 cows per household. That is the real picture of African smallholder dairy: one cow, or two. Not a herd. Not a dairy farm as a magazine would show it. One or two animals standing behind a house, being fed by hand.

That single fact should change how you think about growing this business. If you have two cows and no money to buy a third, then every strategy that begins with "buy more animals" is closed to you. What is left is not small. It is this:

1. Get more days of the year in which your cow is actually in milk.
2. Get more milk on each of the days she is in milk.

Those two levers are the whole of this course. Everything about reproduction, everything about the lactation curve, everything about feeding and calf rearing is really about pushing one of those two numbers up on the animals you already own.

The same Kenyan study also measured what milk does for the household, not just for the cash box. A calving event in the household raised household caloric intake by 11 to 12 percent, and animal-source carbohydrate intake by 36 percent. That is food going into children, measured, not claimed. Adding a calving cow raised household milk yield by 12 to 17 percent. Read those numbers as what they are: measurements from one district in western Kenya over four years, not a promise for every farm. But they show that the value of a dairy cow to a household is partly cash and partly food, and a costing that counts only the litres sold misses part of what she is doing.

Be equally honest about the other side. A dairy cow is a demanding animal. She must be fed every day whether or not she is milking, she must be watered generously, she must be bred back on time, and she can be killed by a disease or an untreated case of milk fever in a day. Milk itself spoils within hours in a hot climate. None of that makes dairying a bad business. It makes it a business that rewards management rather than luck, which is exactly why a course like this can change your income.

So begin with the right question. Not "how many cows should I buy?" but "how many days a year is my cow earning, and how much is she earning on those days?" The rest of this module builds the arithmetic to answer that.

There is one number in dairy that decides more of your income than any other, and most smallholders have never written it down. It is the calving interval: the number of days between one calving and the next.

Here is why it rules everything. A cow only gives milk after she has calved. When that lactation ends she is dry, eating and costing you money and returning nothing, until she calves again and a new lactation starts. So the number of lactations she delivers in her life is set by how often she calves. If she calves often, you get many lactations and many calves out of the same animal, the same shed, the same daily labour. If she calves rarely, you are feeding a cow for months while she earns nothing.

Now the sourced numbers. In that same Asembo, Siaya panel study of 1,317 Kenyan households, the average calving interval was 670 days. The widely used biological target for a well-managed dairy cow is roughly 365 to 425 days. So the measured smallholder reality was nearly double the target, and the gap is not small, it is the whole business.

Work out what that gap costs, in animals rather than money. Take a cow with a six-year productive life.

1. Cow A calves every 365 days. Six years is about 2,190 days. Divide by 365 and she delivers about six calvings and six lactations.
2. Cow B calves every 670 days. Divide 2,190 by 670 and she delivers about 3.3 calvings and 3.3 lactations.
3. Hold the milk per lactation and the value of each calf the same for both cows. Cow B produces on the order of 45 percent less lifetime output than Cow A.

And she does it while eating the same feed for the same six years, occupying the same shed, taking the same daily labour. That is the compounding loss. It is not one missed calf. It is nearly half of everything that animal could have given you, gone, quietly, without any single dramatic disaster you could point to.

This is also why the number is so easy to ignore. A cow with a 670-day interval never looks sick. Nothing goes visibly wrong on any particular day. The loss shows up only when you write two calving dates down and subtract.

The same study gives two more clues about what moves this number. Households with access to communal grazing had calving intervals up to 428 days shorter than those without. Read that carefully. Access to grazing, which is really access to feed, was associated with an enormous shortening of the interval. That tells you the problem is not mainly genetic. It is nutrition and management. The study also modelled that an additional cow in the household was associated with a calving interval about 257 days shorter, which should be read as a correlation from that model, not as a recipe to go and buy a cow.

So what is the target you are aiming at? A cow carries her calf for about 283 days. Add the waiting period after calving before she is bred again, roughly 60 to 85 days if her cycles resume promptly and she conceives, and 283 plus about 80 gives 365 to 425 days. That is where the target comes from. It is arithmetic, not opinion, and the next module takes it apart step by step.

One warning about comparing figures. An improved Boran herd in Kenya has been recorded at a calving interval of 13.8 months, about 420 days. That is a real figure, but it comes from well-managed ranch and research-station conditions, not from a zero-grazing unit behind a house. Do not read it as proof that your breed will do the same on your feeding. Breeds differ, but management differs more.

Everything else in this course, heat detection, feeding in early lactation, body condition, calf rearing, exists partly to pull your own calving interval down toward that 365 to 425 day band. Start by finding out what it actually is.

Plenty of people will tell you what raises milk yield. Very few of them measured it. The Asembo, Siaya panel study did, across 1,317 households and 3,682 household-round observations, and its list of what was actually associated with higher milk yield is short enough to memorise and useful enough to plan around.

Here is the list, as measured:

• A calving cow in the household: household milk yield 12 to 17 percent higher.
• Herds containing exotic or crossbred animals: 6 to 7 percent higher milk yields than indigenous-only herds.
• Acaricide spraying for tick control: associated with a 5 to 11 percent increase in average milk yields.
• Antibiotic use: associated with roughly a 2 percent daily yield increase.

Now read that list properly, because the ranking will surprise you.

First, notice that tick control, which costs a little acaricide and a little labour, was associated with a yield gain of 5 to 11 percent, comparable to or larger than the 6 to 7 percent associated with owning exotic or crossbred animals. Crossbred animals cost a great deal of money to buy. Spraying costs very little. That is not an argument against crossbreeding, which the breeds module takes seriously, but it is a genuine, sourced argument that basic health management is competitive with genetics as a lever, and it is available to a farmer who cannot buy a better cow.

Second, notice how modest the exotic blood effect is in this dataset, 6 to 7 percent, when the breed trials in the next module show crossbreds giving several times the daily yield of a zebu. Both things are true. In a controlled trial with good feeding, exotic genetics deliver a very large yield advantage. In an ordinary smallholder herd fed on whatever is available, the advantage measured across a whole district shrinks to single figures, because genetics only pay when the feeding lets them. A cow bred to give ten litres will not give ten litres on feed that supports three. Remember that whenever someone offers to sell you an expensive cross.

Third, the antibiotic finding needs careful handling. Antibiotic use was associated with about a 2 percent daily yield increase, but that does not mean antibiotics raise milk. Read it as a marker: households treating sick animals are households paying attention to animal health, and it is the attention, not the drug, that most plausibly shows up in the milk. Routine antibiotic use is a bad idea for reasons this course returns to under milk residues, and every antibiotic carries a milk withholding period during which the milk must not be sold or drunk. What that withholding period is for any particular product is set by your national veterinary authority, and this course gives no number in its place. Ask your veterinary officer or medicines authority, and write the answer down.

Fourth, and most importantly, understand what kind of evidence this is. These are associations measured in one Kenyan county across four years. They are not proof that a practice causes a yield increase, and they are not a promise of the same percentage on your farm. What they do is show direction, from real African farms rather than from a research station, and the directions all agree with the rest of this course: health management and management attention pay, and they pay on animals you already own.

So what would you do with this list if you had a small amount of money to spend this month? The evidence points at tick control before it points at genetics. It points at getting the cow back in calf, since a calving cow was the largest single effect in the list, before it points at anything you can buy in a shop. And it points at paying attention, consistently, to animals that mostly look fine.

A last practical note about ticks. The reason tick control shows up so strongly is that ticks do two things at once: they take blood and condition off the animal directly, and they carry tick-borne diseases that can kill her outright. A cow losing condition to ticks in early lactation is the same cow that will fail to come back into heat on time, which lengthens her calving interval, which costs you a whole lactation. The levers in this course are not separate. They pull on each other.

Milk income is a multiplication, not an addition, and writing it out that way tells you where to push.

Income equals litres sold, multiplied by price per litre, multiplied by lactations per year.

Take each in turn, because each one is driven by something different, and only one of them is outside your control.

Litres per lactation. This is set by the animal and by how you keep her. Breed and cross matter, and the range is wide: a Boran-type cow might deliver under 1,000 kg in a lactation, while a well-managed 50 to 75 percent exotic cross can deliver several times that. But breed is only the ceiling. Nutrition and body condition decide how close to that ceiling she gets. Udder health decides whether a quarter is quietly producing less than it should. And the shape of her lactation curve, how well she holds yield after her peak, decides the total more than the peak itself does. Every one of those is a separate module in this course, and every one of them moves this first number.

Price per litre. This is the number you least control and most need to understand. It is entirely local. It differs between selling at the farm gate to a neighbour, selling to a hawker, delivering to a cooperative, and supplying a processor. And, critically, some buyers pay on volume alone while others pay on composition, meaning fat and solids content.

That difference should change your decisions. If your buyer pays a flat price per litre, then a litre is a litre and you are simply chasing volume. If your buyer pays on fat and solids, then a smaller cow producing richer milk can out-earn a bigger cow producing thin milk on the same feed, and a breed choice that looked wrong on litres alone can look right on money. You cannot know which situation you are in without asking. So ask, and write down the answer.

Lactations per year. This is the calving interval from Lesson 2, expressed the other way round. A cow on a 365-day interval delivers one lactation per year. A cow on a 670-day interval delivers about half of one. Because it multiplies the whole equation, improving it improves everything else you have done. There is no point raising your litres per day by careful feeding if the cow then stands open for eight months and you lose the whole next lactation.

Now the honest part. This course gives you no milk price, no heifer price, no feed price and no equipment price, because the reference material for this course contains no verified, dated African price for any of them. Printing a price that was true somewhere three years ago would be worse than printing nothing, because you might build a plan or ask for a loan on it. So the price line is yours to fill in, and it has to be filled in this month, from your own buyers, in your own district.

Here is how to do that properly. Visit or telephone every buyer within reach. For each one write down: the price per litre offered, whether it is paid on volume or on composition, how often they pay and after how many days, what they test on delivery and what happens if the milk fails the test, whether they collect or you deliver, and whether the price changes between the wet season and the dry season. That last question catches many farmers out, because milk prices often fall exactly when everyone's cows are producing most.

Then do one simple comparison. Take your best week of milk records and price that same week's milk at each buyer's terms, including the cost of getting the milk to them and the risk of rejection. The highest price per litre is not always the highest income per week. A cooperative that pays slightly less but collects at your gate, tests fairly and pays reliably every month can be worth more than a hawker paying a better headline price who sometimes does not come.

Write the comparison down and date it. Prices move, and a dated record lets you see next year whether they moved for you or against you.

Most farmers who believe they are making money from milk have only counted the money coming in. Costs in a dairy business are steady, easy to forget, and mostly paid in small amounts that never feel like a business expense. Here is the full list. Write your own figures against each line.

1. Feed. Structurally the largest recurring cost in essentially every dairy system studied anywhere in the world. This is not a claim about a particular price, it is a claim about the shape of the business, and it holds. Everything you cut and carry, everything you buy, everything grown on your own land that could otherwise have been sold, all belongs here. Note that this course deliberately does not teach ration formulation, because there is a separate feed formulation course for that. What it insists on is that you count what the feeding costs.

2. Health and breeding services. Artificial insemination fees or the cost of a bull service, veterinary visits, treatments, dry cow therapy, vaccination and tick control. Note that acaricide belongs here as a cost, and the Siaya study gives you the economic argument for paying it: 5 to 11 percent higher average milk yield was associated with spraying.

3. Labour. Milking twice a day, cutting and carrying feed, watering, cleaning the shed, caring for the calf. This is real time even when it is your own and unpaid. A business plan that treats the farmer's own labour as free will overstate profit, and it will hide the fact that a second cow may be unaffordable in hours even when it is affordable in cash.

4. Housing and equipment depreciation. The shed, the milking bucket, any cooling equipment. Each of these has a working life and then must be replaced. The correct way to count it is to divide what the item cost by the number of years you honestly expect it to last, and charge that amount every year. If you treat the shed as a one-off cost that disappears after year one, your accounts will look good until the day the roof fails and there is no money to fix it.

5. Market access and transport. Getting the milk from your shed to the buyer, every day. Include the cost of any milk that fails a test on arrival or spoils before it gets there.

6. Losses. Discarded milk from a treated cow during her withholding period, milk lost to clinical mastitis, rejected or downgraded milk, and, largest of all and least visible, the milk and the calf lost to an extended calving interval.

That last item deserves its own treatment, because it is where most farmers undercount by the widest margin.

The calf is a co-product, not a by-product. Every extension of the calving interval does not only cost you milk. It costs you a calf. A heifer calf reared properly is either your next milking animal, which means you avoid the cost of buying one, or she is an animal you can sell. In a smallholder system where a purchased crossbred heifer is often the single largest capital outlay a family will ever make, avoiding that purchase is worth real money.

So put a figure on it. Write down what your own next heifer calf is honestly worth to you, either as a replacement you would otherwise have to buy or as an animal you would sell in your own local market this year. That number, not the milk price, is what your calving interval improvements are really protecting, and most farmers have never written it down.

A warning about all of these figures. Every price in this course's source material is unverified, and none is printed here. You are not being given a cost model. You are being given the six boxes to fill and the instruction to fill them with your own current local numbers, dated.

One last discipline. Costs in dairy are lumpy. Feed is paid weekly, veterinary treatment arrives as a shock, a heifer or a chiller is a once-in-years outlay. Cash flow can look fine for months and the business still be losing money, because a big cost has not landed yet. The only defence is to total your costs over a whole year, not a good month.

Nothing in the previous five lessons can be acted on without records. You cannot shorten a calving interval you have not measured, you cannot spot a cow whose yield is dropping if you never wrote down what it was, and you cannot tell profit from turnover from memory. The whole of this course quietly depends on a notebook.

The good news is that the minimum record set needs nothing but paper. Here it is, with the reason each line exists.

Calving date and service date, for every cow. These two dates are the foundation. From the service date you can count 283 days forward to the expected calving date and mark it on a calendar. From the expected calving date you can count back 45 to 60 days and mark the day you must dry her off. From two consecutive calving dates you get the calving interval, which is Lesson 2's number. Without these two dates written down, none of that planning is possible, and remembering is not writing down.

Daily or weekly milk yield, per cow. Not for the whole herd together, per cow. This is the only way to notice a lactation curve dropping faster than it should, or a quiet yield loss from a mastitic quarter before it turns into a visible clinical case. It is also the only honest input to your income calculation. Measure by weight or by a marked container, at the same times each day, and write it at the cow, not from memory in the evening.

Every treatment given: the product, the date, and the calculated end date of the withholding period. This single record is the difference between selling milk with confidence and an accidental residue violation that can cost you your buyer. The withholding period itself must come from your national veterinary or medicines authority for that specific product, because the same active ingredient carries different registered periods in different countries and formulations. This course names no period. It names the record.

Body condition score, once a month. It costs nothing but your hands and five minutes, and it is the earliest warning system you have. A cow losing condition faster than expected in early lactation is telling you about an energy problem weeks before her milk yield does.

Income and costs, however roughly. A monthly total of milk sold and of money spent on feed and health, kept for a year, is what finally answers the question the whole module has been circling: is this cow paying?

Now the honest bottom line, and this course will not pretend otherwise. Nobody can tell you from a book whether smallholder dairying will be profitable for you. That depends on your local milk price, your local feed costs, your local disease pressure, and the specific animal standing in your shed. None of those could be verified for this course, and none is printed here.

What can be said, from sourced evidence, is which levers are real and worth pulling. Closing the calving interval, because a 670-day cow gives roughly 45 percent less lifetime output than a 365-day one. Choosing a breed or cross that matches your feeding and your climate rather than a poster. Controlling ticks, associated with 5 to 11 percent higher yields at low cost. Getting colostrum into the calf. Keeping the udder healthy and the milking clean. Everything else, especially every number with a currency sign in front of it, has to come from your own notebook and your own country's authorities.

So build the notebook. One page per cow, with columns for date, milk morning, milk evening, heat seen, service, treatment given and withholding end date, and body condition score. One page per month for money in and money out. Keep last year's pages, because a record's value is in the comparison. One year of records tells you a little. Three years tells you which cow to keep, which to cull, when your feed gap really falls, and whether the business is growing.

A farmer with three years of records and two cows knows more about their business than a farmer with ten cows and none.