How Many Employees Should I Schedule Each Shift at My Ice Rink?

Signals you actually need this

Most rink operators do not go looking for a staffing formula. The formula finds them, usually after a specific kind of bad month. Here are the signals that tell you your current shift schedule is running on habit instead of data.

Your labor percentage swings more than four points week to week without a matching swing in traffic. Pull twelve weeks of payroll against twelve weeks of revenue. If labor lands at 26% one week and 34% the next while gate counts barely moved, your schedule is not tracking demand — it is tracking whoever asked for hours. A rink with stable traffic and unstable labor cost has a scheduling problem, not a revenue problem.

Somebody is standing around at 2pm and drowning at 7pm on the same day. This is the classic single-rink failure. Freestyle and learn-to-skate blocks in the early afternoon need one guard and a front-desk body. A Friday public session with 180 skaters on the ice needs four or five people spread across rentals, snack bar, ice monitoring, and the box office. If you schedule the same three people from open to close, you are paying for idle hours in the afternoon and eating complaints in the evening.

Skate rental lines exceed six minutes during public sessions. Rental is the choke point at nearly every rink because it is the one station where every single walk-in customer must transact before they can generate any other revenue. A six-minute line at the rental window is a direct measurable loss: some fraction of that line leaves, and the ones who stay skate less and buy less at concessions. If you are seeing that, your issue is almost never total headcount — it is headcount *placement*.

Your managers are scheduling from memory and defending it emotionally. When you ask why Saturday has six people and someone says "that's how we've always done Saturday," you have discovered the actual cost center. Habit schedules ossify around who is available rather than what the floor needs. The revenue-per-employee method exists specifically to take the argument out of the room and replace it with arithmetic everyone already agreed to.

Overtime is appearing on part-time employees. In a facility with a deep part-time bench, any overtime at all is a scheduling defect. It means shifts were built without a running hour count, and someone got backfilled into a fourth or fifth shift late in the week. Rinks with 25 to 50 part-timers should be able to run a full season with effectively zero overtime outside of tournaments.

Your busiest revenue hour and your heaviest staffed hour are not the same hour. This is the single fastest diagnostic. Export hourly sales from your POS for one representative month. Overlay it against scheduled labor hours. If the two curves are offset — and at most rinks they are, usually with labor peaking two hours before revenue does — you have quantified the exact problem this formula solves.

Adjacent tell: your birthday-party and group-booking conversion is dropping. Party bookings are the highest-margin product a rink sells, and they die quietly when the front desk is underwater. If your inbound party inquiries are converting worse than they did last year and nothing else changed, check whether the phone is being answered during the hours parents actually call — typically 4pm to 7pm on weekdays, which is also when your public session rush begins. That overlap is where understaffing quietly destroys the P&L.

What good looks like versus what bad looks like

Bad staffing at a rink is not usually dramatic. It is a slow, expensive drift that looks fine on a shift-by-shift basis and terrible on a quarterly P&L. Here is the contrast in concrete terms.

Bad looks like a flat schedule. Three employees from 10am to 10pm, every day, because that is the number that feels safe. On a Tuesday that is 36 labor hours against maybe $1,400 in revenue — a labor cost around 39% before you touch benefits or payroll taxes. On a Saturday the same three people are handling 300 skaters, the rental window is 12 deep, the snack bar closes early because nobody can staff it, and you leave $600 in concession revenue on the floor while the guards are too busy to actually watch the ice.

Good looks like a stepped schedule that mirrors the revenue curve. Same rink, rebuilt: two people open at 10am for freestyle and stick-and-puck, a third arrives at 3pm as after-school lessons start, two more come in at 5:30pm ahead of the evening public session, and two of the openers cut at 6pm. Same total headcount across the day, redistributed. Labor lands at 24% to 27%, the rental line stays under three minutes, and the snack bar is open through the entire high-traffic window.

Bad looks like counting bodies. Good looks like counting bodies *per station*. A rink is not one job — it is five or six distinct stations: ice monitoring, skate rental and sharpening, box office and admissions, concessions, resurfacer operation, and party host duty. The revenue formula gives you the total; station mapping tells you where they stand. A common allocation on a heavy public session: two on ice, two on rental, one on box office, one on concessions, and a floater who covers breaks and swings to whichever line is longest.

Bad ignores the safety floor. Good treats it as immovable. Under the revenue math, a dead Tuesday morning with $90 in revenue calculates to less than one employee. That output is not usable. Every hour the ice is open to the public requires trained ice monitoring plus someone who can operate the resurfacer and handle rentals. In practice the floor is two people minimum whenever the doors are open, and three whenever more than roughly 40 skaters are on the ice. The formula sets your ceiling responsiveness; the safety floor sets your basement. They are separate systems and you run both.

Bad reschedules reactively. Good reschedules on a cadence. Reactive rescheduling means pulling someone in on Thursday afternoon because Wednesday was busy. That is management by anecdote. A cadence means you pull trailing four-to-eight-week hourly revenue once a month, rerun the division, and republish the template. Weekly tweaks happen for known events — a tournament, a school holiday, a home game — not for vibes.

Bad treats seasonality as a surprise. Rinks have violent seasonality: a September that runs at 40% of a January. If your January template is still live in June, you are burning cash. Good practice is three or four distinct seasonal templates — peak winter, shoulder spring and fall, summer camps and freestyle, plus a holiday-week overlay — each built from that season's own trailing revenue data, not from the annual average.

Real cost and ROI ranges

The reason this method survives contact with a real P&L is that the savings are large, immediate, and easy to audit. Here is where the money actually sits.

Labor is your largest controllable line. At most single-sheet rinks, front-of-house labor runs somewhere between 22% and 35% of gross revenue depending on how disciplined the schedule is. That spread — roughly ten points — is the entire prize. On a facility doing $900,000 a year, moving from 33% to 26% is about $63,000 annually, and none of it requires cutting a single service or raising a single price. It requires moving hours from the dead part of the day into the busy part.

Overstaffing costs are quiet and compounding. One extra person on a five-hour afternoon block, five days a week, at a fully loaded cost of roughly $20 per hour once you include payroll taxes and workers' comp, is about $500 a week and $26,000 over a 52-week year. Nobody notices it as a line item because it is spread across a dozen timecards. It only shows up as a labor percentage that never quite gets where you want it.

Understaffing costs are louder but harder to measure. A rental line that turns away 15 walk-ins on a busy Saturday at a $14 admission plus $6 rental plus maybe $7 of concessions is roughly $400 of lost revenue in a single evening. Do that most weekends through the season and you are looking at real money — plus the reviews, which cost you the following season too. This is the failure mode that scheduling software alone never fixes, because software will happily publish an understaffed schedule if you tell it to.

Tooling costs are modest relative to the prize. Scheduling platforms in this category cluster in two pricing shapes. Per-location tools tend to run in the mid-$20s to around $100 per location per month depending on tier, with free single-location tiers available. Per-user tools typically run a few dollars per employee per month, climbing when you add time-and-attendance and labor forecasting. For a rink with 40 part-timers, per-location pricing is dramatically cheaper — often by an order of magnitude — which is why per-user pricing rarely makes sense for a facility with a deep seasonal bench. Enterprise platforms sold by custom quote exist for multi-site groups and are generally overkill for one sheet of ice.

The payback period is measured in weeks, not quarters. If a $40/month tool helps you shed even four unnecessary labor hours a week, it has paid for itself roughly twenty times over. That is not a marketing claim; it is arithmetic on a $20 fully loaded hourly cost. The real barrier is never the software price. It is getting leadership to actually agree on the per-employee revenue target and then hold the schedule to it.

Where the ROI leaks. Three places, consistently. First, buying a forecasting tool and never connecting the POS feed, which reduces a demand-based scheduler to an expensive calendar. Second, publishing the optimized schedule and then letting managers hand-edit it back toward the old habit shape within three weeks. Third, treating the safety floor as flexible to hit a labor number — which is how you end up with an incident that costs more than a decade of labor savings. The floor is not a variable.

Adjacent economics worth modeling. The same revenue-divided-by-target logic transfers cleanly to comparable venues — trampoline parks, bowling centers, aquatic facilities, and family entertainment centers all share the same shape: heavy part-time bench, violent hourly demand curves, mandatory safety staffing, and revenue concentrated in evening and weekend blocks. If you operate more than one venue type, run the same math on each with its own target number rather than importing the rink's $110 into a bowling center where the per-employee revenue ceiling is structurally different.

How it plugs into your RevOps workflow

Scheduling looks like an HR chore. It is actually a RevOps process, because the inputs are revenue data and the outputs move revenue. Here is how it wires into everything else you already run.

Upstream, it starts at the POS. The whole method depends on clean hourly revenue by day. That means your point-of-sale needs to tag transactions to the right revenue category — admissions, rentals, sharpening, concessions, pro shop, party packages — and timestamp them accurately. If your rental transactions are being rung up in a batch at the end of the session instead of as they happen, your hourly curve is garbage and so is every schedule built from it. Fixing transaction timing is step zero and it is worth a week of retraining.

The target number is a leadership decision, not a formula output. Get your floor managers in a room and settle on the revenue an average employee should support during a normal shift. Say it plainly: "In our building, if you show up, handle a typical crowd, and deliver acceptable service, you are supporting about $110 an hour." That is a floor, not a ceiling. The strong employees hit it doing standard work and then go get the sharpening upsell, the party rebook, the season-pass conversion. Publishing the number means every employee, every manager, and you are all working from the same benchmark.

Downstream, it feeds forecasting and hiring. Once the schedule is built from revenue, your labor forecast becomes a function of your revenue forecast rather than a separate guess. That changes hiring: instead of "we're short-staffed, post a job," you get "peak season needs 340 labor hours a week and we currently have 260 available across the bench, so we need roughly four more part-timers hired by mid-October." Hiring becomes a scheduled activity with a lead time instead of a panic.

Sideways, it connects to training and certification. The safety floor is only real if the people filling it are actually qualified. That means tracking certifications — lifeguard-equivalent ice monitoring training, resurfacer operation, first aid, and whatever your insurer requires — as a scheduling constraint, not a filing-cabinet fact. Several scheduling platforms support credential-based assignment so an unqualified employee simply cannot be dropped into an ice-monitoring slot. If yours does not, you need a manual check in the publish step, every single week.

Compliance is a hard constraint, especially with minors. Rinks run on high schoolers. That brings hour restrictions on school nights, mandatory break rules, and in some jurisdictions predictive-scheduling laws requiring advance notice and premium pay for late changes. Build these into the schedule template rather than catching them in payroll review. A tool with built-in labor-law guardrails earns its price the first time it blocks a 16-year-old from a 10pm Tuesday close.

The weekly operating rhythm. Monday: pull last week's actual labor percentage against actual revenue and compare to plan. Tuesday: adjust the coming template for known events. Wednesday: publish two weeks out so the bench can plan. Friday: confirm coverage for the weekend rush and verify every safety-floor slot has a certified body in it. Monthly: rerun the full revenue division against fresh trailing data and reset the base template. That cadence is the difference between a method and a one-time spreadsheet exercise.

Instrument it so it does not decay. Pick three metrics and watch them: labor as a percentage of revenue by day part, rental line wait time during peak sessions, and unplanned schedule changes per week. The third one is the leading indicator. When hand-edits to the published schedule start climbing, the template has drifted out of alignment with reality and it is time to rerun the math rather than let managers patch it shift by shift.

Related questions

What is a healthy labor percentage for an ice rink?

Front-of-house labor typically lands between 22% and 35% of gross revenue. Well-scheduled single-sheet facilities cluster in the mid-to-high twenties. Anything consistently above the low thirties usually indicates a flat schedule that is not tracking the hourly demand curve.

Should I use the same revenue target for every station?

No. The revenue-per-employee target works for staff whose presence scales with crowd size. Concessions and pro shop staffing should be modeled on transaction volume and queue length instead, since those roles have different throughput ceilings and different revenue attribution.

How far in advance should I publish shifts?

Two weeks is the practical standard for a part-time bench, and some jurisdictions legally require advance notice with premium pay for late changes. Publishing further out reduces no-shows and swap requests, which are the two biggest sources of last-minute coverage scrambles.

Does this method work for a multi-sheet facility?

Yes, but run the division per sheet and per revenue center rather than facility-wide. A twin-sheet rink with simultaneous league play and public skate has two independent demand curves, and averaging them produces a schedule that is wrong for both.

FAQ

What if my ice rink's revenue varies a lot by season?

Rebuild the template each season from that season's own trailing four-to-eight-week data. Most rinks need three or four distinct templates — peak winter, shoulder spring and fall, summer freestyle and camps — plus a holiday-week overlay. Using an annual average guarantees you are overstaffed in the slow months and underwater in the busy ones.

Do I still schedule employees when an hour's revenue is very low?

Yes. Safety does not scale with revenue. Whenever the ice is open to the public you need trained ice monitoring plus coverage for rentals and resurfacer duty, regardless of what the division produces. In practice that is a two-person floor at all open hours and three once you exceed roughly 40 skaters on the ice.

How do I pick the revenue-per-employee target for my rink?

It is a leadership judgment, not an industry constant. Look at your own revenue per labor hour over a good month, discuss it with your floor managers, and set a number that a competent employee can support during a typical crowd. Publish it. The value comes from everyone using the same benchmark, not from the number being externally validated.

What happens if I schedule to the formula and the night is dead anyway?

Send people home only if your local law and your employment agreements allow it, and only down to the safety floor. Better practice is to build flex into the template — one short shift on each heavy block that can be cut or extended based on a two-hour-out gate count, rather than making cut decisions across the whole crew.

How do I handle tournaments and special events?

Treat them as separate overlays on top of the base template, never as edits to it. Estimate event revenue independently, divide by the same target, and add the resulting bodies to the baseline. This keeps your normal template clean and prevents an event week from permanently inflating what everyone thinks the standard schedule looks like.

Does scheduling software replace the revenue math?

No. Software executes and publishes; it does not decide. A demand-forecasting platform connected to your POS can suggest coverage, but it still needs your target number and your safety floor as inputs. Buying a tool without agreeing on the method first produces a very efficient way to publish the wrong schedule.

Sources

• U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics — https://www.bls.gov/oes/
• U.S. Department of Labor, Wage and Hour Division — Youth Employment and child labor rules — https://www.dol.gov/agencies/whd/youthrules
• U.S. Department of Labor — Fair Labor Standards Act overview — https://www.dol.gov/agencies/whd/flsa
• U.S. Consumer Product Safety Commission — https://www.cpsc.gov/
• Occupational Safety and Health Administration — https://www.osha.gov/
• International Ice Hockey Federation — https://www.iihf.com/
• U.S. Figure Skating — https://www.usfigureskating.com/
• USA Hockey — https://www.usahockey.com/
• U.S. Small Business Administration — managing business finances — https://www.sba.gov/business-guide/manage-your-business

Related on PULSE

• [How Many Attendants Should I Schedule Each Day at My Car Wash?](/knowledge/tl0067)
• [How Many Sales Reps Do I Need to Hire for My Logistics Company?](/knowledge/tl0058)
• [How Many Salespeople Do I Need to Hire for My Car Dealership?](/knowledge/tl0052)
• [How Many Producers Do I Need to Hire for My Insurance Agency to Grow My Book?](/knowledge/tl0015)
• [How Do I Figure Out How Many People to Schedule Each Day and at What Times for My Single Store?](/knowledge/tl0002)