Manufacturing accounts: the cost of one unit is not on any invoice

Hook

Boss, if you run a factory, one question: what does one unit of your product cost? Most people start flipping through invoices — "the material was this price, the workers cost this much a month…" Stop. In manufacturing, the cost is not on any invoice. Buying materials has an invoice, paying wages has a record, the electricity has a bill — but "how much did this one item cost" is something nobody invoices you for. So how do you work it out? Today, manufacturing accounts.

A factory has two more stores than a trading business

Start with the difference. Trading is simple: buy a carton in, sell a carton out. One store: goods. A factory is different, and you have three stores. One, raw materials — bought and not yet used. Two, work in progress, WIP — half made, with the material issued and part of the labour done. Three, finished goods — made and not yet sold. So at year end you do not count once, you count three times. And the one most often missed, and most often argued over, is the middle one: work in progress. At the moment a production line stops, what are the half-made items on it worth? They have already absorbed material, labour and part of the factory overhead — they are no longer raw material and not yet finished goods. Plenty of factories have inaccurate books precisely because of that store.

The skeleton of a manufacturing account

Now the structure. A manufacturing account is really a production line with four layers. Layer one: direct materials. Opening raw materials plus purchases minus closing raw materials equals the material actually consumed in the period. Note: how much you bought does not matter; how much you used does. Layer two: direct labour — the wages of those who actually made that batch. Direct materials plus direct labour equals prime cost. Layer three: factory overheads. This is the crux — factory electricity, water, rent, machine depreciation, maintenance, the supervisor's salary, quality control, cleaning. All of it is cost, and none of it belongs to any single unit (the forty-seventh lesson: some costs simply have no invoice of their own). Layer four: adjust for the movement in work in progress. Opening WIP plus the three layers above minus closing WIP equals the cost of production for the period. Add opening finished goods and deduct closing finished goods and you have the cost of what was sold. Look carefully: all the way down that line, "how much was bought" is never the answer. "How much was consumed" is.

Allocating overheads: the traditional way

So layer three, the factory overheads — how do they get spread over each unit? The traditional way is simple: find a volume measure to spread them over. The three most used are labour hours, machine hours, or the direct labour cost. For example: this month's factory overheads came to a certain amount, and the machines ran a certain number of hours across the factory, so you work out the rate per hour; product A ran so many hours and picks up that many hours' worth. When is this good enough? When your products are broadly similar. The same material, the same process, made batch after batch — spreading by hours gives a small error, is good enough, and is cheap.

When the traditional method lies to you: ABC

But what if your products differ a great deal? Here is a very typical situation. You make two products. A is a big order — ten thousand pieces at a time, the machine starts and keeps running, one mould change a month. B is a small order — two hundred pieces at a time, but the customer has many requirements: a mould change every time, machine setup, inspection, separate packing. What happens if you allocate by machine hours? B runs very few machine hours, so it picks up only a little overhead, and on paper B's margin looks beautiful. But the truth is that the people changing the moulds, the setup time, the quality control effort and the store movements are mostly consumed by B. So you read the report and say, "B makes money, take more B orders" — and the more you take, the more you lose. That is the problem activity-based costing, ABC, solves. The idea is simple: do not spread everything with one ruler. First look at which activities in the factory actually consume money: mould changes, setup, raising purchase orders, receiving goods, quality control, packing, despatch. Then for each activity, find the thing that genuinely drives it, the cost driver: mould changes are allocated by number of changes; purchasing by number of orders; quality control by number of inspections; despatch by number of shipments. By count, not by hours. Apply that and products like B — small batch, many variations — reveal their true face at once.

So should you do ABC?

Honestly: not every factory needs it. Three conditions make it worth doing. One, overheads are a large proportion — high automation, many machines, big support departments. Two, the products differ a lot — batch size, process and customer requirements all different. Three, you will genuinely use it to decide something — repricing, cutting a product line, deciding whether to take an order. When is it not worth doing? Products that are all alike, overheads that are a small proportion, or — most commonly — you have not yet got the three stores counted accurately. The order cannot be scrambled: get the three stores counted right and the direct materials and direct labour attributed correctly first, and only then discuss how the overheads are spread. The line from the forty-seventh lesson applies here too: detailed enough to decide on, not detailed without end.

To close

In summary: a factory has three stores and work in progress is where the trouble is. The skeleton is direct materials plus direct labour equals prime cost, plus factory overheads, adjusted for work in progress and finished goods. How much was bought does not count; how much was consumed does. Overheads spread by hours when the products are similar, and by activity under ABC when they are not — otherwise small-batch, many-variation products will pretend to be very profitable. But count the stores first, and discuss allocation second. Want to know which of your product lines is actually subsidising another? Grab a coffee first and talk about your business. For the accounting, come to LTT. I am LTT, helping SME bosses get their accounts straight. Follow us, and see you next time.