Overtime, Agency, or a Hire
Your banquet department runs 22 servers and needs 30 on peak Saturdays. Last month it logged 320 overtime hours - concentrated in the same six servers at $18.50 base, so $27.75 an hour - plus 110 hours of agency labor billed at $31 an hour with a three-hour minimum per shift. Agency shifts actually average 2.4 hours of work; one agency server comped four entrees last month after misreading a dietary card. A full-time server costs $18.50 an hour with a 26% burden and can be scheduled 173 hours a month. The captain is asking for two more full-time heads; the controller says overtime and agency are cheaper than benefits. Everyone is waiting for your answer.
1. Which staffing lever actually wins? Show the comparison.
Read the worked answer
Price each source per worked hour, not per billed hour.
- Overtime: $27.75 worked. 320 hours = $8,880, of which $2,960 is premium.
- Agency: 110 worked hours / 2.4 = about 46 shifts, each billed at the 3-hour minimum = 138 billed hours x $31 = $4,278, i.e. $38.89 per worked hour. The minimum, not the rate, is where the money goes.
- Full-time: $18.50 x 1.26 = $23.31 an hour x 173 hours = $4,033 a month.
One full-time hire covers 173 hours. Take 110 from agency and 63 from overtime: the current cost of that coverage is $4,278 + (63 x $27.75 = $1,748) = $6,026 a month. The hire costs $4,033. Savings of roughly $1,990 a month, about $23,900 a year, per head - and you stop buying untrained labor at $38.89 an hour.
So the controller is wrong on the arithmetic, and the captain is right on the number but for the wrong reason. The case for hiring is not 'we are short-staffed'; it is 'the agency minimum and the overtime premium are paying for hours you never receive.'
2. Who will this hurt, and how do you implement it without losing your best people?
Read the worked answer
The six servers carrying the overtime will experience the hire as a pay cut. Three hundred and twenty hours of premium is real household income, and they have built their lives around it. Handle it explicitly or you will lose your most experienced people in order to save $24,000.
- Tell them before the hire is announced, not after. Overtime will be capped at 40 hours; here is what replaces it - guaranteed full-time schedules, a published two-week shift bid, cross-training into captain and bar roles.
- Fix the root cause, not just the symptom. The overtime exists because Saturday demand is met by stretching weekday staff. A published schedule with a shift bid and no-show tracking costs nothing and reduces the premium hours you are buying.
- On quality: the comped entrees came from an agency worker reading a dietary card nobody trained him on. Any agency use that survives must require a 15-minute site briefing and be confined to non-technical roles - bussing, water, clearing. Never plating, never an allergen plate, never a bar.
3. Which four metrics would you report monthly to prove this worked?
Read the worked answer
- Overtime hours as a percentage of total hours - target under 8%.
- Agency hours, and the billed-to-worked ratio - target under 1.1. A high ratio means the minimums, not the rate, are the problem.
- Cost per worked hour by source - the only comparison that is honest across overtime, agency, and full-time.
- Covers per labor hour on Saturdays - because the point was never cheaper hours, it was the same output for less money.
If agency hours fall but overtime holds, you hired into the wrong shift and the savings are imaginary.