Table of contents
A rota rarely trips over the law and almost always over the calculation before it. The legal framework fits into five rules — a daily limit, an averaging around it, a minimum rest period, fixed breaks, and what applies to Saturdays and Sundays. The harder question is how many sellable hours emerge from your headcount at all.
Six technicians are not capacity. Capacity is what is left after holiday, public holidays, sickness, training and unproductive time — and the difference between those two numbers is regularly larger, in a garage, than a whole member of staff. Hang your rota on the headcount and you plan the same gap in every year.
This article goes through the framework first, because it sets the limits, then the capacity calculation, then the season, then the models that genuinely work in a garage — and finally what upends the plan every week.
The framework in five rules
| What you look up | Why it governs your rota |
|---|---|
| Maximum daily working time | the daily limit, plus the period over which a longer day has to be averaged back down; that averaging is the room in which every alternative model is built |
| Breaks | from what length of work how much break is due, whether it may be split, and how long you may work continuously without one |
| Rest period | how many uninterrupted hours have to sit between the end of one shift and the start of the next — this is the rule that gets in the way of four-day weeks and late Saturdays first |
| Saturday | whether Saturday counts as an ordinary working day, and whether the limit is reckoned per day or per week; that difference decides whether a six-day week is possible at all |
| Sundays and public holidays | when work is permitted, whether a permission is needed for it, and what compensatory rest has to go with it |
The fourth rule is the one most often misunderstood. If your limit counts per day, a six-day week of eight hours — forty-eight hours — is arithmetically possible; if it counts per week, it is not. Look that up before you roster it. The other way round the same applies everywhere: work twelve hours on a Friday and you have a problem a short week does not cure, because an averaging rule is about the average and not about the day itself. And whether you stayed inside your limits is something only the record of hours actually worked shows.
If recovery or roadside assistance is part of it
Then it gets complicated in its own right. Whether your Sunday work falls under an exception or needs a permission, and what compensatory rest day has to go with it within what period, is something to settle with your trade association or the workplace-safety regulator — that is not a question a blog article answers. All that is fixed is the principle: Sunday work is compensated virtually everywhere, and there is a floor on the number of Sundays that have to stay free.
Six technicians are not capacity. Capacity is what is left after holiday, public holidays, sickness and setup time.
Capacity is calculated in hours
The calculation has two stages. First from headcount to attendance hours, then from attendance hours to sellable ones.
| Item | Where the figure comes from |
|---|---|
| Working days per year | with a five-day week, around 260 |
| less public holidays | the recognized holidays where you are; they only count when they fall on a working day |
| less holiday | your contractual holiday; the statutory minimum differs by country and your own agreement usually sits above it |
| less sickness | your own figure from last year, not an industry average |
| less training and school | planned course days, and for apprentices the school and course days |
| times the hours per day | gives the attendance hours per person |
| times the productive share | measured, not estimated: the difference between attendance time and time booked to jobs |
Why the last row is the most important
A worked example with round numbers: 260 working days less 10 public holidays, 25 days of holiday, 10 sick days and 5 course days gives 210 attendance days. At eight hours that is 1,680 attendance hours. If your productive share is 60 per cent, that is 1,008 sellable hours; if it is 75 per cent, 1,260. The difference of 252 hours amounts, in this model, to more than thirty working days — per person.
That is why the question of what share of your hours is productive at all is not a controlling game but the only number connecting the rota to reality. And it has to come from your business: search times, setup times, interruptions, walking to the store and moving vehicles differ more between two garages than between two industries. What the sellable hour ultimately has to cost for the sums to work is what the labor rate calculator settles.
The season is plannable, because it is the same every year
Few trades have an annual pattern as predictable as the motor trade. The holiday check before the summer, the first frost, the bunching of test dates in particular months, and among the customers running two sets the changeover in spring and autumn — that is not only in statistics, it is in your own job figures from the past three years.
From that follows the most effective planning lever there is, and it costs nothing: a holiday plan for the whole year that stands in January. Not because you want to forbid holiday, but because in January all the requests still fit alongside one another and in September not one of them does.
A holiday block is not automatically permitted
With holiday, the employee's wishes count unless serious business interests or socially weightier wishes of others stand against them. So a blanket block on all busy weeks is not something you simply put up — it needs a defensible justification and becomes far less problematic when it is known in January and the remaining weeks are handled generously.
Two more points before the peak. Take plannable work out of the season by actively putting it in the quiet weeks — a customer who has their service in August instead of October relieves you twice over. And plan the peak with attendance rather than with overtime: overtime in season is hours that come back in November as a time bank with no agreement about working it off, exactly when you need them least.
Related articles
Models that work in a garage
| Model | When it fits | What it assumes |
|---|---|---|
| Staggered start times | almost always; the cheapest intervention there is | one person doing intake early while the rest are already working |
| Flexible hours around a core period | where the work is not tied to a takt | a clear core period, reliable time recording, cover at the counter |
| Four-day week | as an argument in the fight for staff | days up to your maximum with the averaging that goes with it; and being reachable for customers on five days |
| Genuine two shifts | where lifts are the bottleneck and not people | enough jobs, plus intake and parts supply in the second shift as well |
| Saturday duty on rotation | where customers drop off and collect on Saturdays | a fixed rotation known in advance instead of allocation at the last minute |
The most common mistake with two shifts: it gets hung on the technicians, not on the lifts. If there are three lifts for six people, a second shift produces something immediately. If there are six lifts for six people, it only moves working time around — and produces a second shift with no intake, no parts counter and nobody to ask.
Two side conditions: for apprentices under 18 stricter limits apply that rule out many models, and you look those up before you roster. And where some form of employee representation exists in your business, it generally has a say in the start and end of daily working time, its distribution across the week and in overtime — in a business of this size that rarely arises, but it is not something you repair afterwards.
What upends the plan every week
A rota planned to a hundred per cent is worthless by Tuesday — the buffer for that belongs in how appointments are made, not in the running week. Four disruptions come back again and again, and for each there is a provision cheaper than firefighting.
- Parts do not arrive. Antidote: sort availability before you give the appointment, not on the morning of the repair. An appointment without a part is a blocked slot.
- Diagnosis takes longer than planned. Antidote: schedule diagnostic time separately, not as the run-up to a repair whose scope nobody knows yet.
- The customer does not come or does not collect. Antidote: confirmation the day before; and for vehicles left standing, a clear rule about when storage time starts being charged.
- Somebody is off. Antidote: a fixed unplanned portion of the day — realistically a good hour per person — and a fallback job always ready that needs no parts.
Standing for a long time is a payment question too
A repair job occupying a lift for weeks ties up not only capacity but money — parts are paid for, work is delivered, the invoice only comes at the end. For jobs like that a deposit and an interim invoice are the appropriate instruments; agree both in writing and set them off visibly on the final invoice.
And the plan itself belongs where the jobs are. A rota on the wall and appointments in another system means two people keeping two truths — and that nobody ever checks the capacity calculation from the second section. How the two come together is what the garage software shows.
The legal framework is quickly dealt with and is rarely the problem: a daily limit with the averaging around it, a minimum rest period, fixed breaks, Sunday only on a proper basis. The actual planning instrument is the capacity calculation — and within it the one number you have to measure rather than estimate: the share of your attendance hours that genuinely lands on jobs. Know that number and you plan the busy weeks in January instead of in September.
Running the garage without paperwork?
Jobs, labor units, parts and invoices in one system. From drop-off to the e-invoice, with nothing entered twice.
