Table of Contents
“The manufacturer pays me four hundred and eighty a case. The technician spends two hours and twenty-four minutes, and that costs me six hundred and twenty-four. I worked at a loss for eight years and called it a partnership.”
Oles Dziuba runs an authorised service centre in Uman. Washing machines, fridges, dishwashers — two brands with official authorisation, plus any commercial repair that comes through the door. Fifteen years mending appliances, the last eight as the owner. Six technicians, his own intake desk, a store of spare parts. And a margin he could never lift.
A workshop of 4 800 visits
Over the past year the centre took in 4 800 visits. That is a little under twenty a working day.
Of those, 2 880 were warranty jobs, appliances still under the manufacturer's warranty. The other 1 920 were commercial, where the client pays himself.
The manufacturer's logic is clear and fair enough in its own way. He works out the average difficulty across the whole network, divides his budget by the number of cases and arrives at a tariff. For a big city workshop with a flow of identical faults that tariff works.
The manufacturer pays per case, the technician works by the hour
A warranty repair is paid for by the manufacturer, not the client, and at the manufacturer's own tariff. The tariff is fixed: 480 ₴ a case. Not per hour, not per difficulty, not per model — per case.
The technician, meanwhile, spends hours. And those hours vary.
Uman is a telling place for a centre like this. A hundred thousand people, two big appliance markets, and the nearest authorised workshop for the same brand a hundred and twenty kilometres away. The queue at Oles's door is not there because he is the best, but because he is the only one.
Two different queues
Physically it is one workshop and the same people. Economically it is two different businesses sharing one resource.
- warranty visits: 2 880
- commercial visits: 1 920
- total: 4 800
- of those, ending with no repair: 640
For years Oles took warranty work to be the base and commercial work the extension on top. The arithmetic says the opposite, and the difference between those two views turned out to be expensive.
Sixty percent of the flow against forty, and Oles always thought that a good ratio. Warranty means predictability: the flow does not depend on whether people have money, because the manufacturer pays.
What the manufacturer pays
The 480 ₴ tariff covers the technician's work. Parts are reimbursed separately, at cost.
Warranty parts carry no markup
On a warranty part the centre earns nothing: it is received at purchase price, fitted and written off. All the margin on parts lives in commercial orders alone, and that is worth remembering before the table.
So out of a warranty case the centre receives exactly 480 ₴ and nothing else.
“The manufacturer asks me once a quarter: why is your average time per case higher than the network's? And I ask him: why is your tariff the same for Kyiv and for Uman, where I wait three days longer for the part?”
What one hour costs
The centre's own cost of a standard hour is 260 ₴. That is the technician's wage with taxes, his share of the premises, tools, diagnostic software and depreciation on equipment.
The figure is not high. Oles worked it out twice and got the same answer both times.
Inside those 2.4 hours the repair is not the biggest part. The waiting is. A part comes from the manufacturer's store in anything from three days to two weeks, and all that time the machine occupies floor space and the client telephones. Every one of those calls is standard minutes too — nobody simply records them.
The difficulty of a case does not move the tariff
A warranty case takes 2.4 standard hours on average: intake, diagnostics, ordering the part, waiting, the repair, the test, the handover. Sometimes an hour, sometimes six, but the yearly average holds steady.
2.4 hours at 260 ₴ is 624 ₴ of cost.
It should be said that 260 ₴ for a standard hour is neither high nor low, it is simply the truth about this centre. In a big city it would be higher because of rent; in a two-man workshop lower, because there is no intake desk. Comparing it with somebody else's is pointless: it describes exactly this set-up, with six technicians and a store of its own.
The warranty case in the red
A tariff of 480 ₴ against a cost of 624 ₴ gives minus 144 ₴ on every warranty case.
Over a year that is 2 880 cases, which is minus 414 720 ₴.
“I got that figure and went off to check whether I had the hours wrong. Put a man with a stopwatch on it for two weeks. The same two point four came out. I was hoping I had made a mistake and it was less.”
So what the centre loses on warranty work is not parts and not carelessness. It loses because it sells hours at a price it did not set, and those hours are longer than the tariff allows for.
The commercial order
Here everything is different. The client pays an average of 1 450 ₴ for the labour, and a commercial order takes 1.8 standard hours, because there is no waiting on a manufacturer's part and no paperwork.
1.8 hours at 260 ₴ is 468 ₴. The contribution comes to 982 ₴ an order.
Over a year, 1 920 orders give 1 885 440 ₴.
There is one more difference that is easy to miss. A commercial client arrives with a fault he has already accepted as his own problem, so he does not argue with the verdict and does not demand a replacement machine. That saves not money but nerves and hours at the intake desk.
Spare parts
Parts turnover in commercial orders is 3 840 000 ₴, with a contribution of 864 000 ₴. That is 22.5% of turnover, or 450 ₴ an order.
Altogether a commercial order brings 1 432 ₴: 982 ₴ of labour plus 450 ₴ on the part.
Parts are the one place where the centre earns calmly. A markup of 22.5% of turnover raises no questions with clients, because a part from the official store with a guarantee on it really is worth more than the same part off the market.
The diagnostics nobody pays for
There is a third category, and it is the most galling of them.
640 visits a year end with the verdict “no fault found”. The appliance works, the client was using it wrongly, or the problem was in the socket, or it went away by itself.
A zero in the “done” column does not mean zero hours
The manufacturer pays nothing for those cases — they are not warranty work, because there was no repair. The client does not pay either, because from where he stands nothing was done. And the technician spent 1.2 standard hours on the intake, the connection, the mode checks and writing up the verdict.
1.2 hours at 260 ₴ is 312 ₴. Across 640 visits that is 199 680 ₴ a year.
Thirteen percent of all visits bring in nothing. That is not sloppiness and not a mistake by the technicians — it is the normal share for domestic appliances, where half the visits open with the words it is making a funny noise.
The year in figures
| Line | Amount, ₴ | Comment |
|---|---|---|
| Warranty income | 1 382 400 | 2 880 × 480 ₴ |
| Commercial labour income | 2 784 000 | 1 920 × 1 450 ₴ |
| Parts income | 3 840 000 | |
| Total turnover | 8 006 400 | |
| Fixed costs | 1 740 000 | premises, 6 technicians, intake desk, software |
| Net profit | 395 040 | 4.9% of turnover |
Four point nine. For a workshop with its own parts store, an authorisation and a queue that is very little, and Oles knew it — what he did not know was why.
The report itself is put together correctly. The income is real, the costs are actual, the profit is genuine. It simply has no line to show that part of the work was done for free, because to the bookkeeping unpaid work is just wages.
How much the centre gives away
| Source of contribution | Amount, ₴ |
|---|---|
| Commercial labour | 1 885 440 |
| Spare parts | 864 000 |
| Warranty cases | −414 720 |
| No-fault-found diagnostics | −199 680 |
| Total contribution | 2 135 040 |
The two negative lines together come to 614 400 ₴.
Unpaid work costs 614 400 ₴ — 1.56 times the centre's entire annual profit.
Per visit that is 128 ₴. The figure looks trivial, and that is exactly why it drew no attention for eight years: nobody goes hunting for a problem that costs a hundred and twenty-eight hryvnias.
Unpaid work has no line in the report
No document anywhere carries a heading for work nobody paid for. The warranty certificates are closed, the manufacturer paid his tariff, and no-fault-found diagnostics generate no document at all. The cost of the technicians sits on one wages line, and there is no way to see which part of it went nowhere.
- warranty case: 2.4 standard hours, tariff 480 ₴, cost 624 ₴
- commercial order: 1.8 standard hours, invoice 1 450 ₴, cost 468 ₴
- no-fault-found diagnostics: 1.2 standard hours, paid for by nobody
- a standard hour in the centre costs 260 ₴ no matter who is paying
Nor is this a story about a bad manufacturer. His tariff is built on the average difficulty across the network, and Uman does not fit that average because of the logistics of parts. Neither side here is lying — one is counting a network, the other is paying for one particular workshop.
What a fair tariff would be
The break-even tariff for this centre is 624 ₴, which is exactly the cost. The manufacturer pays 480 ₴.
The shortfall is 144 ₴, or 23.1% of what is needed.
“I wrote to the manufacturer. The answer was polite: the tariff is uniform across the network, it is reviewed once a year, and your costs are your own responsibility. And that is true, they owe me nothing.”
The right tariff would give the centre 414 720 ₴ a year. It is the one place in the whole model where the decision is not the owner's.
Nor is the tariff a matter for negotiation. It is the same across the network, and a workshop in Uman has neither the volume nor the arguments to change it. It is the frame you have to work inside, not a variable you can steer.
Why none of it is visible
- warranty and commercial orders are closed in the same software under the same document type
- standard hours are nowhere compared with the manufacturer's tariff
- no-fault-found diagnostics produce no financial document whatsoever
- the technicians' wages are one line, with no split between paid and unpaid work
To see the 614 400 ₴ you have to set the actual hours next to what was paid for them. Those two figures live in different systems: the hours in the job log, the payment in the certificate with the manufacturer.
Oles pulled all four lines together by hand over two weeks, because there is nowhere to take them from automatically. That was the real labour of the whole story — not the calculation, but gathering data that lay in three different places.
Giving up the authorisation
The obvious conclusion from the figures: hand back the authorisation and work on commercial jobs alone.
The saving would be exactly 614 400 ₴ — both negative lines disappear along with the warranty flow.
“I had already written the letter. It lay on my desk for two days, and I kept looking at it and thinking: where do my commercial clients actually come from?”
He checked it against two years of history rather than a feeling. He took the warranty visits from the year before last and looked for the same people and the same addresses among the commercial orders of the year after.
Warranty is not income, it is a channel
The answer turned out to be simple. Somebody brings in a machine under warranty, sees the workshop, the technicians, the shelves of parts — and two years later, when the warranty has run out, comes back as a paying client. Or brings a neighbour.
Oles went through the visit history and counted: 25% of warranty clients later become commercial ones.
Two numbers must not be confused here. The conversion is counted off warranty visits, while the contribution is counted off commercial orders, and those are different sets at different prices. Which is precisely why the threshold cannot be guessed by eye.
The 25% turned out lower than Oles had expected. He had assumed about half, because that is how it feels from the intake desk: people come back, say hello, bring their relatives. In reality one in four returns, and that is only enough because the threshold is lower still.
The conversion threshold
Twenty-five percent of 2 880 is 720 commercial orders.
Each brings 1 432 ₴ of contribution, so together 1 031 040 ₴.
| Giving up the authorisation | Amount, ₴ |
|---|---|
| Saving on warranty work and diagnostics | +614 400 |
| Loss of the converted orders | −1 031 040 |
| Effect over a year | −416 640 |
Conversion decides whether authorisation is worth it
The loss comes out 1.68 times larger than the saving. So the authorisation is loss-making in itself and at the same time pays for itself through the flow it brings.
The threshold is easy to work out: 614 400 ₴ divided by 1 432 ₴ gives 430 orders, which is 14.93% conversion. The centre actually runs at 25%, so the headroom is over ten percentage points.
The authorisation makes sense as long as at least one warranty client in six comes back as a paying one.
“The strangest part was realising that both answers are right. Warranty work is eating me alive — true. Without warranty work I close — also true. For eight years I held only the first half in my head, and twice I nearly sent that letter because of it.”
What Oles did
He never sent the letter. Instead he did several things, and not one of them involved the manufacturer.
- no-fault-found diagnostics became chargeable: 350 ₴, credited against the repair if there is one
- warranty and commercial orders are now closed under different document types
- the technicians' standard hours go into every order, no matter who is paying
- the conversion of warranty clients into commercial ones is counted every quarter and hangs on the workshop wall
Headroom of over ten percentage points is not much. If the conversion slides nearer to fifteen, that headroom all but disappears, and Oles will see it on the board in the workshop rather than in a report a year later.
The first point gave the most and hurt the least. Clients did not walk away: paid diagnostics credited against the repair sounds fair, and it cut off the people who used to bring a machine in just to have a look at it.
The fourth point turned out to be the most important. Until then the conversion was an assumption; now it is a figure, and it is that figure which decides whether to stay authorised next year.
The money does not vanish — you simply do not see it
“For fifteen years I thought warranty work was the stability and commercial work the luck. It turned out the other way round: commercial work feeds me, and warranty work brings the commercial in and takes part of the earnings away. Both statements are true, and you have to live with both.”
Nothing changed in Oles's centre: not the equipment, not the staff, not the manufacturer's tariff. Still six technicians, still about five thousand visits, the tariff still 480 ₴. The only thing that changed is that it is now visible how many hours went nowhere, and how many paying clients are brought in by the flow that loses money all by itself.
How to see this in your own workshop
Everything above rests on one comparison that appears neither on the manufacturer's statement nor in the till: how many standard hours went into a case, and how much was paid for it. While warranty and commercial jobs sit on a single line called work completed, the owner sees a 4.9% margin and does not know that 614 400 ₴ of work was given away — while losing even more without it. When hours, tariffs and the conversion of warranty visits into commercial orders are counted automatically, authorisation becomes a decision with a number behind it rather than an obligation — which is exactly why owners of small and medium businesses use Finmap: to see the financial picture rather than piece it together after the fact.
Read also
- Gadget Repair Shop: Real Margin on Parts and Labour
- Auto Service: Which Bay Actually Earns and Which One's Just Busy
Topic foundation: Management Accounting Explained: What It Is and Why Owners Need It
Share valuable content — become a source of insights
Frequently asked questions about authorised service finances
Why can a warranty repair lose money?
Because the manufacturer pays per case while the technician works by the hour. The 480 ₴ tariff is fixed regardless of difficulty, and the average warranty case takes 2.4 standard hours at 260 ₴, which is 624 ₴. Minus 144 ₴ on every case gives 414 720 ₴ over a year.
How do you work out the cost of a standard hour?
Add the technicians' wages with taxes, their share of the premises, tools, diagnostic software and depreciation on equipment, then divide by the actual pool of hours. In this centre it came to 260 ₴, and that figure is the same for warranty and commercial work alike.
Who pays for diagnostics that end without a repair?
Nobody, unless you make them chargeable. The manufacturer does not pay, because the case is not warranty work; the client does not pay, because nothing was done. In this centre 640 such visits a year at 1.2 standard hours came to 199 680 ₴ of unpaid work.
How much work does a service centre give away for nothing?
Add the loss on warranty cases to the cost of no-fault-found diagnostics. Here that is 414 720 ₴ plus 199 680 ₴, together 614 400 ₴ — 1.56 times the centre's entire annual profit. Per visit it works out at 128 ₴.
Is it worth giving up the authorisation?
Work out first how many paying clients it brings in. Giving it up would save 614 400 ₴, but the centre would lose 720 converted orders at 1 432 ₴ of contribution each, that is 1 031 040 ₴. The net effect is minus 416 640 ₴.
At what conversion does authorisation pay for itself?
Divide the total of the unpaid work by the contribution of one commercial order. Here 614 400 ₴ over 1 432 ₴ gives 430 orders, which is 14.93% of 2 880 warranty visits. The actual conversion is 25%, so the authorisation pays for itself as long as at least one client in six comes back paying.
Why does none of this show up in the profit report?
Because the actual hours and the money received for them live in different systems: the hours in the job log, the payment in the certificate with the manufacturer. The technicians' wages sit on one line, with no split between paid and unpaid work, and no-fault-found diagnostics produce no financial document at all.
What can you change if the manufacturer's tariff is not negotiable?
Three things, and not one of them depends on the manufacturer. Make diagnostics chargeable with the fee credited against the repair, close warranty and commercial orders under different document types, and count the conversion of warranty clients into commercial ones every quarter. It is that conversion which decides whether to stay authorised.
Any questions left?
We are ready to answer them.
Finmap support
