Restaurants don't run like a normal 9-to-5 business. They run in fragments: lunch service, dinner service, late-night clean-up, split shifts with unpaid gaps in the middle, different people on different rates, weekend premiums, holiday shifts.
If the payroll system you're buying was built for office workers, it usually breaks the moment you point it at a restaurant.
We wanted to know whether Chegatta's engine would hold up against the real shape of restaurant work. So we built a fictional restaurant scenario and ran it through the engine — no shortcuts, no hard-coded results.
The fictional restaurant
We created Casa Verde Restaurants, Lda. — a fake company in Porto, PT, NIPC 999999990.
Two sites:
- Casa Verde — Porto Centro
- Casa Verde — Matosinhos
Four fictional employees, four different hourly rates:
| Employee | Role | Rate |
|---|---|---|
| Carlos Silva | Waiter | EUR 10.00/hour |
| Maria Santos | Waiter | EUR 11.00/hour |
| Ahmed Khan | Kitchen assistant | EUR 10.50/hour |
| João Costa | Chef | EUR 13.00/hour |
The point of four rates: the engine must not assume every employee gets paid the same. Payroll has to calculate independently for each person.
The real shape of restaurant shifts
Lunch: 11:00–15:00. Dinner: 18:00–23:00.
A waiter might work both. That's a split shift: two working periods with an unpaid gap in between. The gap matters, because payroll systems that simply sum clock-in to clock-out end up paying for the break.
We built scenarios that hit the hard parts:
- Normal split shifts, with break periods explicitly recorded.
- Weekday schedules.
- Overtime after daily and weekly thresholds.
- Night work.
- Weekend work.
- Holiday work.
- Meal allowances.
- Approved leave.
- Unapproved absence.
- Late arrivals.
- Early departures.
The test wasn't: "can the engine produce a number?" It was: "can the engine produce the right number for each employee independently, from this messy week?"
What the engine had to get right
For a restaurant payroll to be trustworthy, the system has to answer questions like:
- How many paid hours did Carlos actually work after excluding breaks?
- When did overtime start — on the daily threshold, the weekly threshold, or both?
- Did the unpaid gap between 15:00 and 18:00 get excluded from hours?
- Did Maria's EUR 11.00 rate leak into João's EUR 13.00 calculation?
- Were the rejected or unapproved records excluded from pay?
- Did the holiday shift get treated correctly?
- Were meal allowances applied only on the days they were earned?
None of these are exotic edge cases. This is a normal restaurant week.
What we proved
We ran the scenario through the engine and confirmed three things:
-
Regular earnings calculate correctly for standard weekday schedules under the test company. The baseline works.
-
Split shifts calculate correctly. The engine summed the working periods, split regular vs daily overtime at the right threshold, and excluded the unpaid break between 15:00 and 18:00.
-
Employee records stay isolated. Changing Carlos's rate or configuration did not affect Maria's, Ahmed's, or João's payroll. Zero cross-contamination. For a restaurant with 20 or 50 staff, that's not a nice-to-have — it's the difference between payroll you can trust and payroll you have to re-check by hand.
The full test passed with 3 out of 3 scenarios green, and it didn't disturb any existing payroll tests.
What "restaurant payroll" really means in practice
For a restaurant owner or a staffing agency placing staff into restaurants, the payroll system has to handle:
- Different hourly rates per employee, with rates that can change over time.
- Split shifts with unpaid gaps.
- Daily overtime thresholds, not just weekly.
- Meal allowances on the days they're actually worked.
- Holiday shifts and holiday premiums.
- Approved leave not turning into an absence violation.
- Unapproved absences becoming compliance events.
- Tip handling as a separate concept from base pay.
- Client billing rates separated from employee pay rates.
- Historical rates preserved, not silently overwritten.
That's the bar. If a system can't clear it, it won't survive a real restaurant payroll run.
Where Chegatta is today
Chegatta's engine clears the core restaurant bar in our tests: split shifts, multiple hourly rates, independent per-employee calculation, overtime thresholds, break exclusion, meal allowances, leave vs absence handling.
What we're honest about: restaurant-specific tip pool distribution and automated multi-step payroll adjustment approval workflows are still optional extensions on top of the core engine. The core calculation and locking are solid. The richer restaurant-specific workflows are the next layer.
If you run restaurants, hospitality, or place staff into restaurant sites, the right question to ask any payroll vendor is: "show me your restaurant week, not your office week." This is ours.