"I build thirty-four kitchens a month and the report shows a profit of 374 000 ₴. But when the installers ring me from site and say it is two centimetres out, I already know: that order never existed for me" — that is how Bohdan, owner of a custom furniture workshop in a regional capital, opened the conversation.
Sound familiar? The orders are there, the deposits come in, the workshop is booked a month ahead, the accountant shows a perfectly normal margin. And there is never any free cash, and every new machine stays a plan for another year.
The reason is almost always the same. In made-to-order manufacturing, money is not lost where it gets counted. The report has materials and it has wages — but the workshop's main resource, the production hour, appears nowhere. And until the hour becomes a line with a price of its own, you cannot see how many hours a month you are paying for twice.
In this article we break a furniture workshop down into two units: the production hour and the site survey. And we show why a two-centimetre mistake costs 13 150 ₴, while the survey that catches it costs 900 ₴.
The founder's road: from one kitchen in a garage to a 400-square-metre workshop
Bohdan started on his own: he ordered the cutting from a neighbour, assembled in his father's garage and did the installation in his own van. He costed the first kitchens on paper — materials, hardware, how much he wanted to earn.
Four years in he had a workshop and his first hired joiner. Today it is 400 m², six people in the workshop, three installers, a dedicated surveyor and a designer.
That is exactly when the clarity disappeared. Volume grew so fast that the workshop had to move twice, and the profit figure in the report looked calm every month. But the feel for the money was gone for good.
"In the garage I knew everything about every kitchen: what I put into it and what I took out of it. Now I have thirty-four kitchens a month and I do not know a single one."
Why "thirty-four orders a month" tells you nothing about money
An order looks like money: there is a contract, there is a sum, there is a deposit. On paper it is all linear — the more orders, the more you earned.
But an order is not money. It is the right to spend workshop capacity. A month holds exactly as many production hours as it holds, and each one is spent once: either on a new kitchen or on reworking an old one.
What the owner sees in the report
Revenue of 2 312 000 ₴ a month at an average order of 68 000 ₴. Materials — 32 000 ₴ per order, 1 088 000 ₴ in total. Workshop costs 850 000 ₴. Total 1 938 000 ₴, profit 374 000 ₴, margin 16.2%.
Perfectly normal numbers for custom furniture. Not one of them explains why the workshop lives in permanent firefighting while the bank account is empty every month.
Life before the site survey became a line
The state Bohdan lived in for almost five years will be familiar to any workshop that grew out of a single garage.
- The production hour was never counted at all. There was one lump sum of workshop wages — and no idea what an hour on the floor actually costs.
- Rework was written off to "the human factor." Not to a separate line with a number against it, but to mood and tiredness.
- The survey was a service, not a control point. The surveyor drove out, wrote down the dimensions — and that was where his responsibility ended.
- Client changes after cutting were handled "like decent people." Free of charge, because "come on, it is just one shelf moved."
- Installer idle time was never recorded. If the schedule slipped, the crew simply waited, and that showed up nowhere.
- Order cost was reduced to materials. 32 000 ₴ per kitchen — and after that "the rest will cover itself somehow."
Not one of these problems was a production problem. All six were accounting problems — and that is precisely why neither the foreman nor the accountant could see them.
The main unit: not the order, the production hour
The first thing Bohdan did was work out what one hour of his workshop costs.
How to calculate the production hour
Take every workshop cost for the month and divide it by the number of production hours. For Bohdan that is 850 000 ₴ over 1 000 hours — which comes to 850 ₴ an hour. That is exactly what the workshop spends every hour, regardless of what is happening inside it right now.
| Workshop cost line | Per month | Per production hour |
|---|---|---|
| Workshop wages (6 people) | 294 000 ₴ | 294 ₴ |
| Surveyor and designer | 96 000 ₴ | 96 ₴ |
| Installers (3) | 141 000 ₴ | 141 ₴ |
| Rent on 400 m² of workshop | 88 000 ₴ | 88 ₴ |
| Equipment leasing and tooling | 62 000 ₴ | 62 ₴ |
| Utilities and consumables | 54 000 ₴ | 54 ₴ |
| Office and marketing | 115 000 ₴ | 115 ₴ |
| Total | 850 000 ₴ | 850 ₴ |
Note this: 319 ₴ out of those 850 ₴ are not people standing at a machine. They are rent, leasing, utilities, office and marketing. They keep running even while the workshop is rebuilding a kitchen that has already been paid for.
How many production hours live inside one order
Divide the workshop's capacity by the number of orders: 1 000 hours over 34 orders — 29.4 hours per kitchen. In money, 29.4 × 850 = 24 990 ₴.
Now the whole order is visible at once. Revenue 68 000 ₴, materials 32 000 ₴, production hours 24 990 ₴. What is left is 11 010 ₴ — and that is the profit on the kitchen, 16.2% of the contract.
Why 11 010 ₴ is the order's entire cushion
Eleven thousand out of a 68 000 ₴ contract is a normal figure for made-to-order manufacturing. The problem is elsewhere: that is the whole safety margin the order has.
11 010 divided by 850 is 13 hours. Thirteen hours over, and the kitchen's profit is gone. In a workshop that does not count the hour this happens quietly: the kitchen is handed over, the client is happy, the acceptance note is signed, the money is not there.
The survey mistake you only see at installation
The mechanics are always the same. The surveyor arrives at a site where the refurbishment is unfinished and takes the dimensions off the bare wall. The designer draws up the project, the client signs it off, the workshop cuts the panels. And at installation the recess turns out to be two centimetres narrower once the tiles are on, and the worktop will not go in.
The worst part here is not the centimetres. It is the moment at which the mistake becomes visible. While the project is on screen, fixing it costs five minutes of the designer's time. After cutting it is a finished part with the edge banding applied, and there is nowhere to put it.
The three line items a rework is made of
- Materials all over again. Board, edging and hardware for the new part — 4 800 ₴. The old board is already cut and gets written off.
- Production hours. Cutting, edge banding, drilling, assembly — 7 hours. At 850 ₴ that is 5 950 ₴, taken out of the next order.
- A second installer visit. The crew, the van, the drive there and back — 2 400 ₴.
Altogether one rework costs the workshop 13 150 ₴. For comparison: the profit on an order done without a single mistake is 11 010 ₴.
An order with a rework against an order without one
Let us put two kitchens side by side. The same contract value, the same client, the only difference being two centimetres at the survey.
| Metric | No rework | With a rework |
|---|---|---|
| Contract value | 68 000 ₴ | 68 000 ₴ |
| Materials | 32 000 ₴ | 36 800 ₴ |
| Production hours | 29.4 | 36.4 |
| Hours in money | 24 990 ₴ | 30 940 ₴ |
| Second installer visit | — | 2 400 ₴ |
| Total costs | 56 990 ₴ | 70 140 ₴ |
| Order result | 11 010 ₴ | −2 140 ₴ |
| Margin | 16.2% | −3.1% |
An order with one rework did not "earn less." It took 2 140 ₴ of the workshop's own money — and another 36.4 hours that could have gone to a new kitchen.
Why a rework eats not part of the margin but all of it
Because the price of a rework — 13 150 ₴ — is higher than the profit on the order, 11 010 ₴. Any rework automatically pushes the kitchen into the red, no matter how well it was sold.
That is exactly why the owner does not understand his own money: he adds orders and the margin does not add up. Some of the orders work with a minus sign, and in the report there is nothing to tell them apart from the rest.
Nineteen percent of orders — and 78 900 ₴ every month
Bohdan pulled up a quarter of history and counted how many kitchens had to be reworked. It came to 19% of orders — six out of thirty-four every month.
Six reworks at 13 150 ₴ each is 78 900 ₴ a month, or 21.1% of the monthly profit. That sum appears nowhere in the report: the materials went into general purchasing, the hours into the payroll, the visit into fuel.
"I used to think a rework was when a joiner ruined a part. It turned out a rework is when we drove away from the survey twenty minutes earlier than we should have."
Installer idle time: the schedule slip that is in no report
A rework is never local: it shifts the schedule. And in installation the schedule is people who have already set off.
Bohdan counted the slips for the month: there were 14, averaging four hours of crew idle time each. That is 56 hours in total, or 14 × 4 × 850 = 47 600 ₴.
Why one slip costs four hours
The crew arrived on site on schedule — and the kitchen is still in the workshop, because the previous one was being finished instead. Moving on to the next order is not an option: that client is expecting his own day. All that is left is to wait.
The workshop pays for those four hours at full price — 850 ₴ an hour — because rent, leasing and wages do not shrink just because the crew is standing in the yard.
Client changes after cutting
The third line item looks the most harmless of all. The client rings back and asks to move the drawers around or add a section — after the cutting is done.
For the workshop it is the same rework at the same price: new materials, hours, a second visit. Bohdan has four such cases a month — 4 × 13 150 = 52 600 ₴. All four were done free of charge, "so as not to spoil the relationship."
"When I saw that free changes for clients were costing me 52 600 ₴ a month, I understood why for five years running I have been the nicest furniture maker in town and the poorest."
A 900 ₴ survey and a 13 150 ₴ mistake
Now the most interesting comparison of all. A surveyor's visit with a verification check costs the workshop 900 ₴: the specialist's time, the van, the drive.
The mistake that survey failed to catch costs 13 150 ₴ — 14.6 times more. All thirty-four surveys in a month together come to 900 × 34 = 30 600 ₴, while reworks caused by surveys and changes come to 78 900 + 52 600 = 131 500 ₴. Divide 131 500 by 30 600 — the workshop is economising on something 4.3 times cheaper than the consequences.
"For years we tried to squeeze two surveys into one day so we would not run the van for nothing. What actually happened was that I handed back the 900 ₴ I had saved in thirteen thousand."
What the workshop actually earned
Let us add the month up. Profit per the report — 374 000 ₴.
| Hidden line item | Amount per month |
|---|---|
| Reworks from survey mistakes (6 × 13 150 ₴) | 78 900 ₴ |
| Installer idle time from schedule slips (56 hours) | 47 600 ₴ |
| Client changes after cutting (4 × 13 150 ₴) | 52 600 ₴ |
| Total hidden | 179 100 ₴ |
| For comparison: profit per the report | 374 000 ₴ |
| Profit after the hidden items | 194 900 ₴ |
Three line items, not one of which appears in any report, eat 47.9% of the monthly profit. The workshop's real margin is not 16.2% but 194 900 divided by 2 312 000 — 8.4%. Over a year the hidden items come to 179 100 × 12 = 2 149 200 ₴.
What 179 100 ₴ is in production hours
Divide the sum by the price of an hour: 179 100 over 850 is 210.7 production hours every month. A fifth of the workshop's entire capacity.
Put differently, one kitchen in every five is made twice. Not because the workshop works badly, but because the mistake becomes visible at installation — when the material is already cut.
Why the surveyor is not to blame
The owner's first reaction is always the same: fire the surveyor. Bohdan did exactly that — and two months later had the same 19% of reworks with a new person.
Because the problem is not attentiveness. It is that the survey is nowhere pinned down as a control point with a price. As long as an hour costs nothing, there is no difference between "re-measured the recesses after the tiling" and "wrote down the dimensions off the bare wall": both options are equally free for whoever is holding the tape.
The moment the production hour gets a price, the difference becomes visible — and nobody needs to be policed by hand any more.
What changed when the survey became a line
Bohdan did three things. Every order got its actual number of production hours and a separate "rework" line with a reason. A verification survey after the finishing work became a mandatory stage priced at 900 ₴ in the quote. Changes after cutting became chargeable at the actual price of a rework.
Two quarters later this is what showed up:
- Reworks from surveys dropped by more than half. A 900 ₴ visit catches what used to surface at installation for 13 150 ₴: a 14.6-fold difference pays it back on the very first case.
- Free changes ended by themselves. When the client sees the price of a change after cutting, half the changes turn out to be unnecessary and the rest get paid for.
- Installer idle time fell along with the reworks. Fewer schedule slips means fewer of those 56 hours the workshop was paying 850 ₴ for and getting nothing.
Not one of these decisions needed new equipment or new people. What they needed was to see the production hour as a line — and to understand that a rework costs more than the profit on an order.
"I spent five years expanding the workshop so I could make more kitchens. What I needed to do was simply stop making one in five of them twice."
The takeaway for business owners. In any made-to-order manufacturing, money is lost not on materials but on repeat work — and the price of a mistake depends on the stage at which it is spotted. So the key number here is not the margin on an order but the price of a production hour and the cost of the control point that catches the mistake before the material is cut. It works the same way in windows, metalwork and printing: a cheap check on the way in is almost always tens of times cheaper than a rework on the way out.
A few closing pieces of advice
- Calculate your production hour. Divide the month's workshop costs by the number of production hours. It is the most honest answer to the question of what any delay costs you.
- Pull rework out into its own line with a reason. As long as it is smeared across purchasing and payroll, it does not exist.
- Compare the price of a rework with the profit on an order. If the rework costs more than the margin, growing volume makes no sense: you are multiplying minuses.
- Make the verification survey a mandatory stage with a price. 900 ₴ in the quote looks like an expense until you set it against 13 150 ₴ at installation.
- Fix the point of no return in the contract. After cutting, changes are chargeable at the actual price — otherwise you are giving clients 52 600 ₴ a month as a gift.
- Count installer idle time in hours. A crew waiting in the yard costs exactly as much as a crew that is working.
- Do not look for someone to blame, look for the invisible stage. If the figure did not change after you changed the person, the point is that the stage is nowhere weighed in money.
Money does not disappear — you simply cannot see it
Custom furniture looks like a business where everything is decided by materials and the joiner's hands. In fact that is only the ticket in, and the money is decided by one thing: how many production hours you spend on work you have already been paid for.
None of these costs goes away on its own. But the moment the production hour becomes a line with a price, and the survey a control point with a sum, you can see where you earn, where you do the work twice — and what you can do about it as early as next week.
How to see this in your own workshop
Everything described above rests on a single table: every order is a line showing the price, the materials, the production hours consumed and, separately, the hours spent on rework. Most owners keep this in the foreman's notebook and total it up at the end of the month. When the same data is collected automatically, you can see which order ate twice the time it was priced for — which is exactly why small and medium business owners use Finmap: to see the financial picture instead of reconstructing it from memory.
Read also
- Printing cost: how to calculate press hour and makeready
- Cost per labour hour in cleaning: where the margin goes
Topic foundation: Management Accounting Explained: What It Is and Why Owners Need It
Frequently asked questions
Add up everything the workshop costs you in a month — wages for the production team, the surveyor and the designer, the fitters, rent, equipment leasing, utilities, the office and marketing — and divide it by the production hours actually worked. In the example in this article that is 850 000 ₴ over 1 000 hours, so 850 ₴ an hour. Materials are not inside that figure: they are counted separately on every order.
Because materials are the smaller half of the cost. An order worth 68 000 ₴ carries 32 000 ₴ of materials, and it looks like 36 000 ₴ is left. But the order eats around 29 workshop hours, and that is another 24 990 ₴. What actually stays is 11 010 ₴ — 16.2%, not 53%. The gap between those two numbers is exactly the time nobody counted.
13 150 ₴: material again 4 800 ₴, seven production hours to build the unit a second time 5 950 ₴ and a repeat trip for the fitters 2 400 ₴. For comparison, the site survey itself costs 900 ₴. So the mistake is 14.6 times more expensive than the work that was supposed to prevent it. With rework, the order brings not 11 010 ₴ of profit but −2 140 ₴.
Because it has no line of its own. The material is written off as an ordinary expense, the workshop hours are not logged anywhere, the second trip looks like part of the fitters' normal work. Until rework is booked against a specific order with the reason written next to it, it just looks like the cost wobbling a little.
In the example 19% of orders — six out of thirty-four — need at least one part remade. That is 78 900 ₴ a month. Add the fitters standing idle because the schedule slipped (47 600 ₴) and the free changes clients ask for after cutting (52 600 ₴), and you get 179 100 ₴ — almost half the workshop's profit.
Split them by the moment they arrive. While the board has not been cut, a change costs a few minutes of the designer's time — do it for free, it pays for itself. After cutting it costs exactly what rework costs, because physically that is what it is. That line has to be in the contract and said out loud to the client in advance, not at the moment he picks up the phone.
No, it is the most expensive saving in custom manufacturing. All the site surveys in a month cost 30 600 ₴, while rework and post-cutting changes cost 131 500 ₴ — four times more. A second visit to a difficult flat, a check survey after the old furniture comes out and photos of the pipes and wiring pay for themselves with the very first rework you avoid.
Start logging time per order instead of the workshop's overall load. In the example 1 000 production hours across 34 orders average out at 29 hours. Once those hours sit on specific orders, you can see that the complicated projects swallow twice the average while the price list charges the same for them.
