In this article
A split shift reads like an 8-hour day on the rota. It is actually a 12-hour day for the person living it — and the hours in the middle belong to nobody: too short to go home and do anything useful, too long to just wait around, and in nearly every European market, covered by no law at all.
Almost every full-service restaurant in Europe runs on the same shift shape without ever naming it: someone works the lunch service, clocks out, and comes back for dinner. It has a name in every one of this site's 24 markets — coupure in French kitchens, turno partido in Spanish ones, turno spezzato in Italian ones, geteilter Dienst in German ones, gebroken dienst across the Low Countries — and it is one of the most ordinary facts of hospitality life anywhere the same kitchen serves two peaks a day.
This blog has already priced the gap the law actually regulates: the rest period **between** two calendar days, the EU's 11-hour floor under Article 3 of Directive 2003/88/EC, and what a close-then-open "clopening" shift does to it. That is a different gap from this one. This article is about the hole **inside** one calendar day — lunch out, dinner in — and it is a completely different legal and economic animal.
No EU directive touches the same-day gap at all. The Working Time Directive sets maximum weekly hours, minimum daily and weekly rest, and a break during a long shift — but it says nothing about how many hours may separate two working periods on the same day, or whether they must be paid. Where a limit exists, it lives in national labour law or in a hospitality collective agreement, and it differs from country to country and often from region to region within one.
So this article does something different from a legal explainer: it treats the gap as an arithmetic problem, the same way this site's drinks-margin and benchmark tools treat cost and revenue. Seven numbers, two graphics, and a live calculator at the end that turns your own shift times, your own days per week and your own hourly cost into your own answer — not an EU average that may not apply to you at all.
Why the Gap Is the Real Cost, Not the Hours
Ask anyone who has worked a split shift for a season why they eventually left hospitality for retail, logistics or an office, and the gap comes up before the pay does. It is too short to genuinely go home in most cities once a commute eats into it twice, too long to spend productively standing around the pass, and it makes a second job, an evening class, a school pickup or simply a normal errand nearly impossible to plan around. It is the single most commonly cited reason for leaving a kitchen or a floor that is not about the wage on the payslip at all.
It is also the cost an owner never sees on a report, because nothing about it generates a line item. Payroll shows hours worked and hours paid; it has no column for hours consumed. A rota built entirely around split shifts can look perfectly efficient — low idle-labour cost, tight staffing against covers — while quietly being the least sustainable schedule the business runs, because the cost it is hiding shows up as turnover eighteen months later, not as a number this month.
And it is genuinely not the same question as the EU rest-period article on this blog. That one asks whether there is enough time **between** today's closing shift and tomorrow's opening one — a real legal floor, 11 hours, that either is or isn't met. This one asks what happens **inside** today, between a lunch that ends and a dinner that starts, where there is no floor to check against at EU level at all — only arithmetic, and whatever your own country or collective agreement happens to say.
The 7 Numbers Behind the Gap
Run these on your own rota, not on the worked example below — the calculator two sections down does exactly that. But the shape of the problem is the same everywhere a lunch and a dinner service share one employee.
1. Amplitude Is the Number That Matters, Not Hours Worked
Payroll thinks in hours worked. The person living the day thinks in amplitude — the wall-clock span from the moment they leave home for the lunch shift to the moment they get home after the dinner shift. In the worked example below, that's 11:00 to 23:00: a 12h day built around 8h30m of actual work.
The four-hour difference between those two numbers is not a rounding error. It is the entire subject of this article, and it is the number a rota that only tracks scheduled hours never shows anyone — not the owner running payroll, and not the person whose whole day it consumes.
Lunch service, the gap, dinner service — drawn to the true proportion of a 12h day.
The gap alone is 3h30m — 29% of the whole day, and none of it shows up on a payslip.
2. The Dead Middle Is Too Short to Use and Too Long to Ignore
In the worked example, the gap itself is 3h30m. That single number is the whole problem, and it sits in an unusually unhelpful zone: long enough that standing around the pass for it is a genuine waste of a life, short enough that going home rarely works once a commute is subtracted twice — once each way — and whatever is left barely covers a proper meal, let alone a errand, a class or a second small job.
This is exactly the gap that shows up nowhere on a timesheet. It isn't clocked, it usually isn't paid, and unless someone deliberately measures it, it is invisible to everyone except the person standing in the middle of it.
3. The Dead-Time Share Turns the Gap Into a Percentage
Divide the gap by the amplitude and the abstraction becomes a number anyone can feel: 29% of the entire day, in the worked example, is neither worked nor free. Not "time off" in any usable sense, and not paid time either — a third category the rota has no word for.
Framed as a share rather than a duration, this is the number worth watching if you run several split-shift rotas side by side: two schedules with identical worked hours can carry very different dead-time shares depending purely on how the gap sits inside the day, and the share is what predicts which one your staff will tolerate longer.
4. The Same Gap, Five Days a Week, Compounds Into More Than a Part-Time Job
One 3h30m gap is a bad afternoon. Five of them, the normal rhythm for someone rostered onto split shifts most of the working week, add up to 17h30m of unpaid, unworked time — every single week, on top of every hour they were actually scheduled to work.
Multiplied across a working year of roughly 48 weeks, that reaches 840h. Divide it by the length of the person's own worked shift and you get somewhere around 99 "shifts' worth" of pure dead time a year — not overtime, not a bonus, just hours their schedule consumed without paying for or crediting them at all.
Each row is one split-shift day; the 3h30m gap repeats identically, day after day.
By the end of the week that's 17h30m gone — and across a working year, 840h of unpaid, unworked time.
5. Zero — the EU Directives That Touch a Same-Day Gap
That's the actual number: zero. The Working Time Directive regulates the length of the working week, the rest between two calendar days and the break inside a single long shift — it simply does not address how a working day may be split, or whether the gap in the middle must be compensated.
Where a limit or a compensation rule exists, it is written into national labour law or into a hospitality sector collective agreement — never harmonised across the EU. French kitchens talk about la coupure, and the sector's own collective agreements have long addressed daily amplitude and coupure-related allowances. Spanish hospitality's many regional convenios de hostelería regulate el turno partido, typically limiting how the working day may be interrupted. Belgium's PC302 hospitality committee has its own provisions for gebroken diensten. None of those numbers travel — a rule that applies in one country's collective agreement says nothing about the next one, and the details change over time even within a single country. Treat every specific figure you find, including anywhere else on this site, as a starting point for checking your own current national law and your own sector's collective agreement — never as a fact to build a rota on without verifying it.
6. The Coverage-Risk Number: What an Empty Middle Costs When Dinner Opens Cold
The gap is not just a cost to the person working it — it is a coverage risk to the business running it. Someone who leaves for 3h30m and doesn't reliably come back on time (traffic, a second obligation squeezed into the gap, simple fatigue) opens dinner service short, on exactly the shift the business can least afford to run thin.
The two ways owners actually manage that risk both cost something: pay someone to effectively stand by through the gap (a real wage cost for time that produces nothing), or accept the coverage risk and absorb the occasional late arrival, no-show, or scramble to cover a table section at 18:35. Neither is free; the split shift only makes the second one look free, because its cost is a probability rather than a line on an invoice.
7. The Break-Even the Calculator Below Actually Answers
Every number above is worth knowing on its own, but the question an owner actually needs answered is a comparison: what would it cost to compensate the gap properly, against what the gap is quietly worth if you valued it at the same hourly rate as the rest of the shift? That comparison is where a coupure-style allowance either looks cheap or looks like a rounding error next to the real cost — and it depends entirely on your own numbers, not the worked example's.
The Split-Shift Calculator
Type your own lunch and dinner times, how many split days a week the schedule actually runs, your all-in hourly cost (wage plus employer charges) and — if you already pay one — a daily gap allowance. Every tile below recalculates as you type.
There is no legal threshold built into this tool, on purpose: because no EU-wide rule exists for a same-day gap, the honest version of this calculator is pure arithmetic, not a pass/fail compliance check.
Your Own Split-Shift Numbers
Type your real shift times, days per week and hourly cost — every number below recalculates instantly.
—
This is arithmetic, not a legal minimum — there isn't one at EU level. Check your own country's labour law and hospitality collective agreement for any coupure/split-shift allowance you may already owe.
Two numbers are worth sitting with. The annual gap tells you how much of one person's year your rota consumes without paying for it; the allowance comparison tells you how far your own compensation practice — if you have one — actually goes toward covering that value.
If you don't currently pay any allowance for the gap, set that field to zero and watch the verdict change: it will simply state the annual value of the uncompensated time at your own hourly rate, which is the number worth bringing to a conversation about whether the rota needs to change rather than the pay.
What to Actually Do About the Gap
Three honest options, in the order most kitchens reach for them:
If the split shift has to stay
- Shorten the gap wherever the service model allows it — even 30 minutes off 3h30m compounds meaningfully across 5 days a week.
- Pay a standby or coupure-style allowance for the gap, sized against what your own calculator above shows the time is worth — not an arbitrary round number.
- Protect at least one full, genuinely uninterrupted day off around every split-shift block; a week of nothing but split days with no real recovery day is the fastest route to the turnover this article opened with.
- Tell new hires the true amplitude, not just the worked hours, before their first shift — a surprise gap on day three costs more goodwill than a disclosed one on day one.
If the gap can be removed
- Run a staggered, continuous-shift team instead: one crew covers the lunch-into-afternoon, a second covers late-afternoon-into-close, and nobody personally carries the gap.
- Cross-train so lunch and dinner can genuinely be different people on the days volume allows it, rather than the same person twice.
- If the service model supports it, consider all-day dining or a simplified afternoon offer that keeps the kitchen open through what used to be the dead middle — removing the gap rather than compensating it.
The Gap Is Worth Pricing, Even Without a Law to Enforce It
There is no EU number to hit here, no compliance box to tick, and that is exactly why this gap gets ignored: nothing forces an owner to look at it. But the arithmetic doesn't care whether a law requires it — 3h30m unpaid, unworked hours in the middle of a working day is real time consumed from a real person's life, whether or not any authority is checking.
The restaurants that keep split-shift staff longest are rarely the ones paying the most per hour worked. They're the ones who ran this arithmetic once, know exactly what their own gap costs, and made a deliberate choice about it — shorten it, compensate it, or design it out — instead of inheriting whatever shift pattern the previous owner happened to leave behind.
Run your own numbers in the calculator above, then read the rest-between-shifts article for the other half of the same rota: the gap between today's last shift and tomorrow's first one, which — unlike this one — does have a legal floor to check against.