Every October, operators near Zilker make the same mistake. They treat ACL as a demand event and spend their energy on pricing, listings, and visibility, chasing bookings that were going to come anyway. The festival sells out the city. Demand is not the problem. The problem is what happens after the booking, when two weekends of full occupancy hit a fulfillment system built for a slow Tuesday.

The leak is a misdiagnosis. When operators frame ACL as a demand opportunity, they invest in capture and starve fulfillment. Then the bookings arrive, the turnovers stack, the messages flood, the calendar collides, and the operator discovers the real constraint at the worst possible moment. The festival did not test whether you could fill units. It tested whether you could deliver on the units you filled.

Demand work feels productive. Adjusting prices, polishing photos, watching the booking count climb. It is also the part of the business that, during a sellout festival, takes care of itself. Spending September on demand is optimizing the input when the output is the bottleneck. The operators who do this arrive at October 2 with full calendars and empty systems.

Capacity is unglamorous. It is cleaning schedules, message sequences, calendar hygiene, and staffing depth. It does not show up as a number that climbs. It shows up as the absence of disaster, which is exactly why operators underinvest in it until disaster arrives.

Your real capacity is the number of bookings you can fulfill to standard, simultaneously, without anything breaking. It is not the number of units you list. A four-unit operator with a one-person manual process has a capacity of maybe two clean festival turnovers a day. The other two are gambles. Demand will happily exceed that capacity, because demand does not know or care what you can deliver.

When demand exceeds capacity, you do not earn more. You earn the same revenue plus a stack of bad reviews, refund requests, and burnout that costs you the next six months. Overbooking your own capacity is not growth. It is borrowing against your reputation.

A single festival weekend can be survived on adrenaline. One bad night, one missed turnover, you recover. ACL runs October 2-4 and October 9-11, and the second weekend arrives before you have recovered from the first. Adrenaline does not refill in five days. Whatever broke in weekend one, the missed cleaning, the contradictory message, the double-booking, breaks again in weekend two, now with a tired team and depleted supplies. The two-weekend structure is what turns a capacity gap from a bad night into a pattern.

The useful question is not how many bookings can I get. It is how many can I deliver before something breaks, and where does it break first. For most operators the first failure is turnover, then communication, then the calendar. Knowing your ceiling lets you make a real decision: build capacity, or cap bookings at what you can honor. Both beat discovering the ceiling live, in front of paying guests.

Operators who push past capacity during festivals report the same after-math. Revenue that looked strong on paper gets eaten by refunds, comped stays, and review damage that suppresses the next quarter's bookings. The booked number and the kept number are not the same. The gap between them is exactly the size of the capacity you did not build.

Demand is borrowed. It arrives because of the festival, not because of you, and it leaves when the festival leaves. What you keep is the operating layer you build to absorb it: the turnover logistics, the comms sequences, the calendar discipline, the staffing depth, run as one spine. That system serves you every weekend after, not just the festival ones. Demand is the stress test. The capacity to fulfill it is the prize.

If you do not know your real capacity ceiling or where it breaks first, find out before October hands you more than you can hold. The free STR Leak Scorecard locates your fulfillment constraints and ranks them by what fails soonest under festival volume. Own the rails before demand exposes the leaks.