Dynamic Class Pricing: When Off-Peak Discounts Actually Pay Off

Off-peak discounts pay off only when the class is already running under half full and the discount pulls in a visit that would not otherwise have happened. The test is the Marginal Seat Rule: an empty mat earns nothing, and the instructor is paid either way. Everything else is leakage from seats you had already sold.

Does dynamic class pricing actually work for a studio?

It works on classes where the instructor is already contracted and the room is under half full. On sold-out classes it is a straight margin cut.

Picture two reformer classes in the same 20-mat room. Tuesday 6:30pm sits at 18 booked. Tuesday 10:30am sits at 5. Both cost you the same $65 instructor hour and the same rent. Discounting the 6:30pm class costs real money. Discounting the 10:30am class converts capacity you were about to throw away. Demand is not the constraint: gym and studio membership reached a record high in 2025, with more than 80 million Americans holding a fitness facility membership, according to the Health & Fitness Association. The constraint is when that demand lands.

Which classes qualify for an off-peak discount?

Run the Four-Week Fill Test on the individual slot. Under 45 percent mean fill, discount it. Between 45 and 70 percent, market it. Above 70 percent, leave it alone.

Pull four consecutive weeks of the same slot from your booking export. Drop any week containing a public holiday, because a Memorial Day Monday will drag an otherwise healthy slot below the line and you will discount a class that did not need it. Take the mean, not the worst week.

The counterintuitive part: most owners already believe they know the dead slot, and they are usually wrong. Saturday afternoon feels slow because the lobby is quiet, but the actual hole is often Tuesday 1:15pm or Thursday 11:00am. Run the test per slot, per instructor. A 10:30am with one instructor and a 10:30am with another can differ by 15 points of fill, and that is a staffing decision, not a pricing one.

How much do you discount before the class moves?

Use the Off-Peak Ladder. Start at 20 percent, hold three weeks, step to 30 percent, hold three weeks, stop at 40 percent. Never go past 40.

Round the published price down to a clean number every time. A $28 drop-in at 20 percent off is $22.40. Publish $22. Nobody books a class priced at $22.40, and the odd number makes the discount look calculated rather than offered.

Past 40 percent off, the price stops reading as a time-of-day deal and starts reading as a cheaper class, which is a permanent repositioning you did not intend. A discount that fails at 20 percent very often fails at 30 percent too. If two ladder steps add fewer than three bookings per class, price is not the barrier. The time is. That is a schedule problem, and no discount fixes a schedule problem.

How do you stop peak members from sliding into the cheap slot?

Fence the rate three ways: a time fence, a redemption fence, and a booking fence. Fences live in the product, never in a coupon code.

The time fence states the window in plain language on the schedule: off-peak rate applies to classes starting 9:45am to 3:00pm, Monday through Thursday. The redemption fence blocks off-peak credits from being used after 4pm with no front-desk override, because the override always gets used for one nice regular and then for everyone. The booking fence releases off-peak seats 48 hours out instead of two weeks out, so the organized member who would happily have paid $28 cannot pre-plan their whole month into the discount.

Why does per-visit math beat per-member math here?

Because members do not price their own attendance the way spreadsheets assume, and your unlimited members are already paying a high effective rate per visit.

The clearest evidence comes from a much-cited 2006 study of 7,752 members across three US health clubs. Members on plans above $70 a month attended about 4.3 times monthly, an effective cost near $17 per visit, while a $10 per-visit pass was available at the same desk. At the prices of the time, that gap cost the average member roughly $600 over the life of the membership.

The operator lesson is not that members are irrational. It is that an off-peak discount aimed at your own unlimited base subsidizes visits you already sold, at full price, in advance. Aim the off-peak rate at drop-ins, class packs, first-timers, and members who have gone quiet. That last group is where the return sits, and it is worth pairing with at-risk member detection so the discount lands before the cancellation does.

Should you raise the peak price instead of cutting the off-peak one?

You can, but never present it as a peak increase. Raise the standard rate once a year at renewal and let off-peak be the discount off that number.

The spread is identical either way. The perception is not. In February 2024, Wendy's mentioned testing dynamic pricing on an earnings call, and the coverage read it as surge. The company issued a clarification saying it would not raise prices "when our customers are visiting us most," and reframed the plan around offers in slower times of day. The research on price fairness points the same direction: when buyers discover that someone else paid less for the same thing, they judge the seller as less fair and report lower satisfaction, even when their own price never moved. Same math, different frame, very different Instagram comments.

How do you run the six-week off-peak pilot?

One slot, six weeks, no other marketing pointed at it. The Off-Peak Pilot is deliberately boring, because attribution dies the moment you run two variables.

• Week 0. Record the baseline: fill rate, gross revenue, instructor pay, and the names of everyone who attended. You need the names to tell new visits from moved visits.
• Week 1. Publish at 20 percent off. One slot. Not two, not a whole daypart.
• Weeks 1 to 3. Hold. No extra ads, no extra emails, no Instagram push. If you spend against it, you will never know which lever worked.
• Week 4. Step the ladder only if the slot added fewer than three bookings per class.
• Week 6. Decide on contribution per class hour.

Add one step that costs nothing: call every off-peak first-timer within 48 hours of the class, not next week. Ask what they normally book. Half will tell you they cannot make evenings, which is the whole business case in one sentence.

Which numbers tell you the discount paid off?

Contribution per class hour, and peak fill rate, together. Headcount on its own will always make the discount look like a win.

Contribution per class hour is seats sold times net price, minus instructor pay, minus variable cost per head. Net price matters more than operators expect at low ticket values, because card processing takes a larger proportional bite out of a $16 seat than a $28 one; if off-peak becomes a real volume channel, the payment method mix is worth revisiting.

Then check peak. If your 6:30pm fill rate drops more than five points across the pilot while off-peak rises, you moved revenue instead of creating it. Revert immediately and tighten the fences before trying again. Across Zatrovo studios, 2026, failed pilots went wrong this way far more often than by failing to fill the discounted class.

When should you kill the class instead of discounting it?

If six weeks and two ladder steps leave mean fill under 30 percent, the slot is dead. Cancel it and redeploy the instructor hour somewhere with demand.

Moving a Tuesday 1:15pm barre to Tuesday 6:15am is usually a better decision than pricing it at $16 forever. The direct loss is not the real cost of an empty class anyway. A four-person class in a 20-mat room signals decline to every trial member sitting in it, and trials convert on the room feeling alive. Consolidating two half-empty slots into one busy one often raises revenue and retention at the same time, with no price change at all. Rebuilding the grid this way pairs naturally with a review of your membership pricing tiers.

How do you announce it without triggering a fairness backlash?

Sequence it: instructors first, existing members second, public schedule third, roughly 72 hours apart. Never describe it as a price change.

Instructors get asked at the door and will improvise an answer if you have not given them one. Give them a single line: "Daytime classes are $22 because they run under half full, and the evening price is not changing." Existing members hear it next, framed as an addition rather than a restructure. Only then does it go on the public schedule.

Put the off-peak price on the schedule next to the class time, visible before booking. A price that appears at checkout after someone has chosen their spot reads as a trick even when it is a discount. That is the same mechanic behind checkout abandonment, and it applies just as hard to a $22 mat.