Scheduling and capacity: room-hours, turnover, and the peak that sets your ceiling

Ask an operator how full their studio is and the answer usually arrives as a feeling about evenings. That feeling is often correct and is never sufficient, because it cannot be used to decide whether to add a room, extend the hours, sell a corporate account, or accept a partnership that brings twelve people in at once. Those decisions need a capacity model; for a recovery studio it is simpler than it sounds, once the unit is right.

The unit is the room-hour. Not the appointment, not the member, not the square foot. A room, a chamber, a bed or a chair is available for a certain number of hours each day, and everything you sell is a slice of that availability. Member counts describe demand and tell you nothing about whether you can serve it. Two studios with identical membership rosters and different room counts are running different businesses with different ceilings.

Two things then reduce that raw number, and both are commonly left out. Turnover consumes room time that produces no revenue. And demand arrives concentrated, not spread, so most of your room-hours exist at times nobody wants them. A capacity model that ignores either one produces a confident overestimate, which is the specific error behind studios that add members faster than they can serve them and then wonder why the reviews turned.

Counting room-hours: what an independently bookable resource is

Start with the count of independently bookable resources, and be strict about the word independently. Two saunas that share a changing area, a plunge that cannot run while the filtration cycle is on, or three compression chairs in one room that can only be attended as a group are not the number of resources they look like on a floor plan. Count what can genuinely be sold to unrelated people at the same time, which is often fewer units than the equipment list implies.

Multiply by the hours each resource is actually available, which is not the same as your open hours. Some equipment needs a warm-up or a cool-down at the start or end of the day. Shared-water modalities lose time to chemistry checks and to any circulation cycle you will not run with a member in the water. A room used for classes at fixed times is not available for individual bookings during those windows. Deduct all of it at the source instead of pretending it away and then wondering why the schedule never matches the model.

The result is your gross room-hours per day, per week, or per month, and it is the denominator for everything that follows. Compute it per modality, not for the studio as a whole, because a studio is rarely constrained in aggregate. It is constrained on one resource at one time of day, and an aggregate number hides exactly that.

Turnover is a deduction, not a detail

Every session in a shared room has a second component: the time between one member leaving and the next one being able to start. Wipe-down and sanitation, linen, resetting controls, water chemistry on a plunge, a cool-down or reheat cycle on a heat modality, and the simple fact that the outgoing member is still getting dressed. None of that produces revenue and all of it consumes the room.

So the schedulable block is the session plus the turnover, and this is where most capacity models break. If a thirty-minute session sits in a room that needs fifteen minutes of reset before the next person can start, the room is occupied for forty-five minutes, not thirty. Modelling it as thirty overstates that room's capacity by half. Those numbers are illustrative, not measured, and your own session and reset times will differ by modality, but the arithmetic is the point: capacity is computed on the block, and the block always includes the tax.

Measure your own turnover rather than estimating it, and measure it at the busiest hour rather than at ten in the morning, because turnover under pressure with one person on the floor takes longer than turnover with time to spare. Time it repeatedly, use something close to the slow end of what you observe, and write the block length into the booking system as the actual slot rather than leaving staff to absorb the difference by rushing. A slot length that lies to the schedule produces a floor that runs late all evening and a member who waits in a lobby wondering why they booked a time.

A worked example, with inputs you should replace

The following inputs are illustrative and invented for arithmetic. They are not benchmarks, they are not observed anywhere, and the only part of this section worth copying is the sequence of operations.

• Say a studio has four independently bookable sauna rooms and opens twelve hours a day. Gross room-hours are 4 x 12, or 48 per day.
• Say each room loses half an hour a day to warm-up and end-of-day shutdown. Available room-hours fall to 4 x 11.5, or 46.
• Say the session is 30 minutes and turnover is 15. The schedulable block is 45 minutes, so each room supports 11.5 hours divided by a 0.75-hour block, which is about 15 blocks per room per day, or roughly 61 across the four rooms.
• Say demand is concentrated so that four of the twelve hours carry most of it. In those four hours the ceiling is 4 rooms x 4 hours / 0.75, or about 21 sessions, and that number, not the daily 61, is what a busy Tuesday evening is competing for.
• Now compare: modelling the same studio on a 30-minute block with no turnover would have produced about 92 daily sessions and 32 at peak. The overstatement is the entire turnover tax, and it compounds every time you extrapolate revenue from it.

Peak concentration is the number that binds

Recovery demand is shaped by the working day. People come before work, at lunch in some markets, and in the two or three hours after the working day ends, and the after-work window is usually the largest by a wide margin. Weekend demand has its own shape, typically a long morning. The consequence is that a studio can be at a comfortable average and still turn people away every evening for a year.

Compute the peak separately and treat it as a distinct capacity question. Take your busiest recurring window, compute the room-hours available inside just that window, apply the same turnover deduction, and compare it against demand for that window, not against total demand. That comparison answers questions the daily average cannot: whether a new membership tier is sellable, whether a corporate account can be accommodated, and whether the next room you buy should be another unit of your busiest modality rather than a new modality entirely.

The rest of the day is a separate problem, and it is a demand problem, not a capacity one. Off-peak room-hours are inventory that expires unsold every day, and the standard moves against that are the ones that shift demand instead of adding supply: off-peak pricing or access tiers, corporate accounts scheduled deliberately into the daytime trough, partnerships with nearby employers or clinics whose people are free at those hours, and class formats placed where individual demand is thin. corporate-wellness-accounts-for-recovery-studios treats that as its central subject.

Three ways a utilisation number overstates itself

Utilisation is used room-hours divided by available room-hours, and every part of that expression is easy to compute generously. Three specific errors account for most of the overstatement, and all three are worth checking before you act on your own dashboard.

• Counting resources that are not independently bookable, which inflates the denominator on the floor plan and then gets quietly corrected by reality at the busiest hour.
• Counting hours nobody wants. Including every open hour in the denominator produces a low average that reads as spare capacity, when the spare capacity is at eleven in the morning and the shortage is at six in the evening.
• Counting bookings rather than attendance. A booked slot that nobody used consumed the room and produced neither revenue nor a member who felt served. If no-shows are in your utilisation numerator, the number is measuring your schedule rather than your business.

Scheduling rules that buy capacity back

Once the model is right, several ordinary scheduling decisions turn out to be capacity decisions. They cost nothing in equipment and they are worth working through before any conversation about buying another room.

• Set slot length equal to the real block, session plus turnover, so the schedule stops running late and the last booking of the evening stops finishing after close.
• Stagger start times across resources rather than starting every room on the hour, which spreads the load on the desk, the changing area and the showers, all of which are shared even when the rooms are not.
• Sequence modalities deliberately where members combine them, since a contrast sequence occupies two resources in a fixed order and scheduling the second one independently produces a member waiting wet in a hallway.
• Run a waitlist on peak windows and make it automatic, because a late cancellation at the peak is the most valuable inventory in the building and the least likely to be resold by hand.
• Cap first-visit bookings at the peak. First visits take longer and consume desk time, and a peak hour is the worst possible moment to give someone their first impression of the studio.
• Restrict any supervised or high staff-attachment service to windows where the coverage already exists, which is a scheduling change with a payroll benefit.

When the room is not the constraint

It is worth confirming that the resource you are modelling is the one that actually binds, because on many floors it is not. Shared infrastructure is a frequent culprit: changing rooms, showers, a single accessible restroom, or a lobby with four chairs will cap throughput below the room-hour ceiling, and the symptom is a schedule that is theoretically fine and a floor that feels chaotic. Water treatment capacity and the recovery time of a heat source can do the same, since a plunge that cannot hold temperature under back-to-back use is not the resource its spec sheet describes.

Staffing is the other common binding constraint, and it interacts with the room model rather than sitting beside it. Every additional concurrent session adds greeting, screening, turnover and safety obligations, and at some point the marginal session requires a second person on the floor. That is a step change, not a smooth cost, so the profitable capacity moves are the ones that fit under the existing coverage. staffing-a-recovery-studio works that arithmetic from the labor side.

Work out which one binds by watching where the queue forms, not by reasoning about it. If the rooms are full and nobody is waiting, the rooms bind. If the rooms are not full and the lobby is, something shared is binding. If everything is fine except that staff are visibly behind, the constraint is coverage, and adding equipment will make it worse.

Keeping the model alive as the menu changes

A capacity model is a snapshot of a floor plan, and it goes stale the moment anything about the menu changes. A new modality with a different session length changes the block arithmetic. A membership tier that encourages higher frequency changes peak demand without changing the room count. A class format converts individual room-hours into scheduled ones. Recompute after each of these rather than carrying forward a number that was accurate under a previous configuration.

Recompute against attendance, too, not against bookings, and against the peak window, not the day. The pair of numbers worth tracking over time is peak-window utilisation and revenue per available room-hour, both computed from your own records. The second one is what actually tells you whether a modality is earning its floor space, because a resource can be busy and still be the wrong thing to have in that room. What sessions and entry-level memberships are listed at across our own directory is counted below from the listings themselves, and that is the anchor to compare your own revenue per room-hour against, since the price side of that ratio is the part an operator most often sets by intuition.

The two guides that sit either side of this one are worth reading with the model in hand: choosing-your-modality-mix works throughput per hour and capital per station as a portfolio decision, and sauna-throughput-and-turnover works the same arithmetic in detail for the heat modalities, where session length and cool-down interact in ways that surprise people who planned a floor around square footage alone.