Farm Record Keeping Guide

Farm records answer one question that nothing else on a Ugandan farm can answer: which enterprise makes money and which one quietly loses it. This guide sets out the smallest workable system, five records rather than twelve, and shows how they produce a cost per kilogramme you can hold against the price a buyer offers at the gate. Farm record keeping is measurement, not paperwork.

What Farm Records Tell a Ugandan Farmer That Nothing Else Can

A farmer can know the crop grew well, the birds looked healthy and the buyer paid, and still not know whether the season made money. Those three things are all observations. None of them is arithmetic. Farm records turn a season into a number, and that number is the only thing you can set against an offer at the roadside.

The gap matters most where the margin is thin, which in Uganda is nearly everywhere. This site has now costed more than twenty crop and livestock enterprises, and the pattern is consistent: the difference between a profitable acre and a loss making one is often a few hundred shillings a kilogramme. A farmer working from memory cannot see a gap that small. A farmer with a book can.

There is one piece of Ugandan evidence on how common the habit is, and it is worth stating carefully because it is narrower than it looks. A peer reviewed household survey covering 450 randomly selected households across six districts of central Uganda found that 22 percent kept meaningful records. The same study cited an earlier assessment in south western Uganda that put the figure at 14 percent. Both samples were smallholder dairy households, both covered one region, and neither study defined what counted as meaningful. So treat these as two regional readings rather than a national rate.

The association inside that survey is the part worth dwelling on. Among the households keeping records, 24.3 percent used artificial insemination to improve their herd. Among those keeping none, 1.4 percent did. Record keepers were not made better farmers by their books, and the study does not claim they were. What the contrast shows is that the two things travel together, and that a farm which measures itself tends to be a farm that changes something.

Why No Reliable Ugandan Cost Figure Exists for Most Farm Enterprises

Here is the strongest argument for keeping records, and it comes from failing to find data rather than from finding it. Across this site's cost pages, one problem recurred on almost every enterprise: no dependable Ugandan figure exists for what a thing actually costs to produce or what it actually yields. Not because nobody published a number. Because the farmers who could have measured it did not write it down, so every published figure traces back to a budget assumption, a promotional claim or a recall interview.

Several of those pages had to be inverted as a result. Instead of stating a profit, they state a break even figure and let the reader compare it to their own situation, because the input data to do anything better does not exist. That inversion is a workaround for missing records, repeated across a whole content library.

Look at what the arithmetic does when it is done honestly. On maize, the ministry's own per acre budgets put the cost of production between 411 and 647 UGX per kilogramme, while the farm gate price at harvest runs from 400 to 1,000. Read those two bands together and the conclusion is uncomfortable: the cheapest acre in the country is sometimes loss making, and it has no way of finding out.

Bush beans are starker. A fully costed acre needs about 885 kg to break even at 1,300 UGX a kilogramme, 639 kg at 1,800 and 460 kg at 2,500. Uganda's measured national bean yield sits at 263 to 324 kg per acre. Every break even figure in that list is above what the country actually grows.

Livestock behaves the same way. Fattening pigs on bought feed breaks even near 7,200 UGX per kilogramme liveweight against a farm gate band of 4,500 to 7,000, so the pig loses money before it reaches the trader. And a 500 bird Kuroiler batch on purchased compound feed came out thin to negative, while every competing page showed a profit, because those pages used a promotional growth claim of 3.5 kg in ten weeks instead of measured feed intake against roughly 2 kg at sixteen weeks.

Four enterprises, four cases where the sums only work out one way once someone counts properly. A farm with records does not need this page to tell it which side of the line it sits on. A farm without them is guessing, and the guess is usually optimistic.

The Smallest Set of Farm Records Worth Keeping

A guide that asks a smallholder to keep twelve ledgers will be ignored by the second week of planting, and it deserves to be. The ministry's own training manual makes the same point in its design rules for farm records: if a record keeping system is complicated, it is more likely to generate mistakes. So start with the smallest set that answers a real question, and add only when the small set is running without effort.

Five records do the job. What went out, what came in, and against which enterprise.

Notice what is missing. No balance sheet, no annual valuation, no asset depreciation, no weather log. Those are all in the standard lists and all of them can wait. A rain gauge is interesting and a cash book is not going to tell you your cost per kilogramme without the four production records underneath it.

Compare that with the international standard and the five hold up well. The United Nations Food and Agriculture Organization runs a survey design for agricultural holdings that asks whether a holding records its activity or finances at all, then asks what is registered across eight categories: area cultivated or harvested, crop production, livestock production, unit prices and amounts sold and total sales by product, input quantities used, detailed quantities and prices of inputs bought, workers' time, and workers' payment. Every one of the five records above maps onto that list. The categories the five leave out are the ones a farm can reconstruct later from the rest.

Separating Enterprises: Maize, Goats and a Poultry Flock in One Book

If you take one idea from this guide, take this one. Every shilling in and out must carry the name of the enterprise it belongs to. A farm running maize, a few goats and a poultry batch that keeps one cash record cannot tell which of the three is carrying the other two. That is exactly how a loss making enterprise survives for years: it hides inside the takings of a profitable one.

This is not a rare arrangement in Uganda. It is the normal one. Uganda's Annual Agricultural Survey, published by the national statistics office in collaboration with FAO and the World Bank, found that about 95 percent of agricultural households cultivate crops and about 70 percent keep livestock, with nearly every livestock keeping household also growing crops. Roughly seven in ten Ugandan agricultural households run crops and animals side by side, out of one purse.

Separating them costs nothing. It is one extra word on each line. Write M for maize, G for goats, P for poultry, or use a page per enterprise in the same book. What it buys is the ability to close each enterprise off at the end of a season and ask whether it paid. Do that twice and you will find one of them was never worth the land.

Plot separation matters for the same reason. The same national survey found that 51 percent of agricultural households used two crop plots or fewer in a season, 33.8 percent used three or four, and 15.6 percent used five or more, on an average crop plot of 0.7 to 0.8 hectares. So most farms have a small number of distinct, nameable pieces of ground, which makes yield by plot practical rather than bureaucratic. Name them the way your family already names them and use those names in the book.

Recording Farm Inputs Bought, With Quantity and Price

An input record without a quantity is useless for costing. "Fertilizer 120,000" tells you what left the purse and nothing about whether the rate was right, whether the price was fair, or what it cost per kilogramme of grain. The ministry's own input record format asks for six things, and the quantity and unit price are two of them.

Keep the receipt where you can find it, and if the dealer will not write one, photograph the goods and the price board with your phone before you leave. A photograph with a date on it is a record. It is a weaker record than a receipt and a much stronger one than a memory.

Tools and durable items belong in the same book with their expected life noted, which is all the asset inventory a smallholding needs. Once the five records are running you can take stock once a year and see what you own. Trying to run an inventory before you can run a sales record is the wrong order.

Farm Labour Records by Operation, Not by Day

Most record keeping advice tells a farmer to log wages by the day. Ugandan enterprise budgeting does it differently, and the difference is worth copying. The ministry's per acre maize budget prices labour by operation: a figure for digging holes, a figure for weeding, a figure for top dressing, a figure for harvesting, a figure for herbicide application. Not days times a day rate.

Its record format follows the same shape. The rows are tasks, and the columns are when it happened, how long it took, how much labour it used and what it cost.

Two reasons the task version is better. First, it survives a changing wage. Rates move district by district and season by season, so a record that says "weeding an acre took nine person days" stays useful long after the rate that year stops meaning anything, while a record that says "paid 180,000 for labour" tells you nothing you can reuse. Second, it lets you price the job before you commit. Uganda's national survey has to ask farmers to recall an average number of days and an average day rate over a whole season, because that is the most a recall interview can get. Your own book can do considerably better than that, since you are there on the day.

Family labour is the line everyone skips, and skipping it is how a farm convinces itself an enterprise is profitable. If a crop only breaks even because nobody charged for six weeks of family weeding, that is worth knowing before you plant another acre of it. Write the person days even when no money moved. The labour cost calculator can put a value on them afterwards.

Yield Records by Plot, and Why Harvest Weight Beats the Basin

This is where most Ugandan farm records quietly break. The harvest gets written down in basins, tins, jerrycans or unmarked bags, and none of those has a fixed weight. Earlier work on this site tried to publish a standard weight for a tomato box, a basin and a basket and had to refuse, because those units are not standardised in Uganda and no reliable conversion exists.

A record in basins cannot become a cost per kilogramme. That is the whole problem in one sentence. You can divide shillings by kilogrammes and compare the answer to a price. You cannot divide shillings by basins and compare the answer to anything, because the buyer's basin and yours are not the same object.

The evidence that this is a real leak sits inside Uganda's own statistics. The Annual Agricultural Survey states plainly that production data is collected as farmer declared estimates, with farmers reporting harvests in the local units of measure they use, and those quantities then converted into metric tonnes afterwards using available conversion factors. So the national yield figures for this country pass through a conversion step that nobody at farm level can check. If that is how the numbers are made, the farmer who weighs is holding better data than the national average.

The ministry's own storage records already assume a scale. Its stack card is described as a card fixed to a stack of bags keeping a tally of the number and weight of bags added or removed. A farm weighing scale is the single piece of equipment that makes the whole record system work, and it is the one purchase this guide argues for without hesitation.

Record the weight that leaves the plot, plot by plot, and record what the household ate alongside what was sold. Around 61 percent of the maize harvested in a Ugandan second season is sold unprocessed and much of the rest is eaten at home, so a yield record built only from sales receipts will understate the crop badly. What you ate is production. It just did not turn into cash. A crop production sheet kept per plot handles both columns.

Farm Sales Records: Weight, Unit Price and Who Bought It

The sales record is the one most farms already keep in some form, usually as a memory of what the trader paid. Made properly it carries six things: date, product, weight sold, price per kilogramme, the type of buyer, and how they paid.

The ministry's format asks for quantity sold and average price per unit sold, which leaves the unit open. Close it yourself. Write kilogrammes and shillings per kilogramme, and if the trader insists on buying by the bag, weigh the bag before it goes. Two records of the same transaction, one in bags and one in kilogrammes, are worth the extra thirty seconds when the price moves next season.

The buyer column earns its place quickly. Record whether it was a bicycle trader at the farm, a bulker at the trading centre, a miller, or a school. After three or four seasons you will have your own evidence of which route pays best for which product, and that is a stronger basis for pricing your produce than anything a page can tell you. Note the mode of payment too, because a credit sale that was never fully paid is still sitting in most farmers' heads as a sale.

Where one payment covers several products, split it on the spot. A lump payment written as a lump is a lost record. Guessing the split a month later reintroduces exactly the recall error the book was meant to remove. A sales sheet with a weight column makes the split obvious because the weights will not add up otherwise.

Mortality and Herd Records That Decide a Livestock Margin

On a livestock enterprise, deaths are the line that turns a workable budget into a loss, and they are the easiest thing to forget because nobody wants to write them down. A poultry batch is priced per bird placed and paid per bird sold, so mortality acts as a divisor on every cost in the budget. Lose a tenth of the flock late in the cycle and every shilling of feed those birds ate is still in your costs with nothing on the other side.

Record the date, the number, the age or stage, and what you think happened. You are not making a diagnosis. You are building a pattern that a veterinary officer can read in one minute, which is worth more than a description from memory. If deaths cluster in the same week of two consecutive batches, that is a finding, and it will not appear in a running total.

For goats, cattle and pigs, the useful record is a running head count with births, deaths, purchases and sales dated. Goats in particular return herd growth rather than cash for the first years, so a farmer with no head count has no way to see the return at all. Four extra kids on the ground is real income that never passed through a pocket. A livestock record sheet by animal keeps the count and the health notes on the same page.

Working Out Your Cost Per Kilogramme From Farm Records

Here is the payoff. Cost per kilogramme is the only figure that can be set directly against a price offered at the gate, and it falls straight out of the five records with no extra work.

Do it once and you will understand why the cost pages on this site keep inverting into break even figures. Do it twice and you will be able to walk away from a price instead of accepting it because the money is needed today. The farm profit calculator and the break even calculator both take these figures as their inputs, and neither is worth opening until you have them. Current gate prices for comparison sit on the maize price page and its siblings.

One warning about the arithmetic. Add your own columns rather than trusting a printed total in any budget template, including the ones in official manuals. Totals get typed wrong. Yours is the only one you can check.

Record Formats That Survive a Ugandan Harvest Week

The best record system is the one still being written in the middle of a busy harvest week, not the most complete one. That single test settles most of the paper against phone argument.

A hard cover book kept dry does the job. Rule the columns yourself on the first page of each section, keep it in a plastic bag inside the house rather than in the store, and write in pen. Ruled columns beat loose paper because loose paper migrates. One book with the five records in five sections beats five books, because five books means four of them are somewhere else when you need to write.

The phone earns its place as a camera before it earns it as a ledger. Photograph receipts, photograph the scale reading on a full bag, photograph the price board at the agro shop. Those images are dated automatically and they survive a wet walk home better than a paper slip. Transcribe them into the book weekly, because a phone gallery with four hundred photographs is storage, not a record.

Spreadsheets and software are genuinely better once the volume justifies them, and the volume on a smallholding usually does not. FAO's own survey designers reached a similar conclusion from the other direction: where a farming household keeps no complete accounts, their remedy is to hand the household a diary and shorten the period being asked about, then visit four times instead of once. Their fallback for missing records is a simple book filled in as things happen. That is the same instrument this guide is recommending, and it is recommended because it works, not because it is basic.

What Farm Records Unlock Beyond Knowing Your Own Margin

Records open doors, and it is worth being accurate about which ones and how far.

Credit. Most guides say records get you a loan. They do not. A lender decides on security and on whether the repayments can be met, and a book is neither of those things. What records do is let you answer the second question with evidence instead of a hopeful estimate, and let you borrow a sum that matches what the enterprise can actually service rather than what you feel able to ask for. That is a real advantage and it is smaller than the promise usually made. The page on preparing for an agricultural loan goes into what is actually asked for.

Cooperative and group membership. Groups that bulk produce have to allocate proceeds between members, and allocation runs on weights delivered. A member who cannot show what they delivered takes what the group decides they delivered. Farmer cooperatives and aggregation groups are the most immediate practical use of a weight record.

Contracts with buyers. A buyer who needs consistent volume wants to know what you can supply and when, and the only honest answer comes from two or three seasons of yield records. Without them you are quoting a hope, and the penalty for missing a contracted volume falls on you.

Insurance and input support. Be sceptical here. Uganda's Annual Agricultural Survey lists agricultural insurance among the costs a farming household might incur and reports no measurable spending on it at all, printed as a blank rather than a figure. So records may position a farm for a scheme that arrives, and no record will conjure a scheme that is not there. Treat this as the weakest of the four reasons.

Worth noting what all four have in common: they need a record that already exists. Nobody can build a supply history retrospectively, which is the argument for starting the book in a season when nothing depends on it.

Where Farm Record Keeping Usually Fails on Ugandan Farms

The failures are predictable and mostly structural rather than about discipline.

The system was too big at the start. A farmer sets up nine registers after a training, keeps them for three weeks, and abandons all nine. Five records kept badly are worth more than nine kept for a month. Start smaller than feels right.

Nothing gets written at the point of the transaction. Records written in the evening are already recall. Records written a week later are fiction with a date on it. The book needs to be where the money changes hands, or the phone does.

One purse, three enterprises. Covered above, and it stays the most expensive mistake on this list because it can run undetected for years.

Nobody ever adds it up. A book that is written in and never totalled is a diary. The totals are the product. Set a fixed point, the end of each season, and close every enterprise off then.

Only one person on the farm can read the book. The ministry's guidance on operational records makes this point: a farm record entry should be simple enough that everybody working on the farm can understand it. If the system lives in one person's handwriting and shorthand, it dies when that person travels.

Advisory support is thin, which makes self reliance here more than a preference. The national survey found that the share of agricultural households receiving any advisory service ranged from about 4.3 percent to 22.9 percent depending on the sub region. Most Ugandan farmers are not going to be taught this by a visiting officer, so the book has to be simple enough to set up alone.

Frequently Asked Questions

How long should farm records be kept? Keep at least three seasons of the same enterprise, because two seasons give you a comparison and three start to show whether a pattern is real or was just weather. Old books are worth keeping indefinitely once you have them, since they cost nothing to store and become the only long run yield data your farm will ever have. If a book is falling apart, photograph the pages before it goes.

What does it cost to start keeping farm records? Close to nothing for the book. A hard cover exercise book and a pen are among the cheapest items in any agro shop or stationer, and ruling your own columns costs an evening. The one real spend is a scale, and the price depends on the capacity you need, whether you buy a hanging spring balance or a platform type, and where you buy it. Prices vary by supplier and season, so compare locally rather than working from a figure on a page. The scales guide covers what each type handles, and a shared scale between two or three neighbouring farms is a reasonable way in.

Do I need a smartphone or software to keep farm records? No. A phone is useful as a camera for receipts and scale readings, and nothing about the five records needs an application or a data connection. FAO's own answer to missing records on smallholdings is a paper diary filled in as events happen, which should settle the question. Move to a spreadsheet when the volume of lines makes the book slow, not before.

How do I record produce the household ate rather than sold? Weigh it and write it in the yield record with the note that it was consumed, then leave it out of the sales record. It belongs in production because you grew it, and leaving it out understates your yield and inflates your cost per kilogramme. In a country where a majority of several staple harvests is eaten at home, a yield figure built only from sales is badly wrong.

Will keeping farm records get me a loan? Not by themselves. Lending decisions turn on security and on demonstrated capacity to repay, and records speak to the second only. What they do is let you show a real margin and borrow an amount the enterprise can service, rather than guessing. Any page promising that a record book secures credit is overstating it.

What if I cannot read or write easily? Use weights, tallies and marks rather than sentences. A stack card is a tally of bags and their weight, and it needs no prose at all. Photographs of the scale reading, dated by the phone, work as a yield record. Many farms run the book through a school age family member, which works as long as the writing happens the same day and the totals get read back to the person making the decisions.

How do I record labour done by my own family? Write the person days exactly as you would for hired workers, and put the cash column at zero or at the local rate, whichever you prefer, as long as you do the same thing every season. The reason to record it is that unpaid family labour is the most common thing hiding a loss making enterprise. If a crop only clears its costs because six weeks of family weeding was free, you want that on the page before you plant it again.

Start the book this season, on the enterprise you are least sure about. One hard cover book, five sections, an enterprise tag on every line, and a scale you can borrow if you cannot yet buy one. Price a scale with a local dealer or at a farm equipment stockist and check the current gate price for your crop before you commit to a sale. By the end of the season you will own one number nobody else in this country can give you, which is what it actually costs you to produce a kilogramme. Everything else on the tools and planning pages becomes usable once you have it.