Case Study 24.1 — Selling Dinner as a Ticket: Next, Alinea, and the Prepaid Reservation
A real, public case. Facts drawn from public record and the founders' own extensive public statements about their model. Where a figure has not been independently verified, it is described as a claim rather than a finding.
Background
In April 2011, a restaurant called Next opened in Chicago's West Loop. It was the second project from chef Grant Achatz and his business partner Nick Kokonas, whose first restaurant, Alinea, had opened in 2005 and become one of the most decorated in the United States.
Next's culinary premise was unusual — a menu that changed completely several times a year, each one an entire cuisine or era rather than a list of dishes. But the thing that made it a case study for this chapter was not the food. It was the way you got in.
Next did not take reservations. It sold tickets.
You bought dinner in advance, the way you buy a seat at a concert. The ticket carried a date and a time. It was paid for at purchase — tax and service included — and it was, in the general case, non-refundable, though transferable: if you could not come, you could give or sell your ticket to someone else. Prices varied by night and by time slot, with a Saturday at eight costing more than a Wednesday at nine-thirty.
Kokonas built the ticketing software himself because nothing on the market did what he wanted. He later spun it out as a separate company, Tock, which launched publicly in the mid-2010s, expanded to thousands of restaurants and wineries internationally, added a widely used takeout module during the COVID-19 shutdowns of 2020, and was acquired by Squarespace in 2021.
That arc — an operator builds a tool to solve his own problem, the tool turns out to be the more scalable business — is common in restaurant technology. What concerns us is the problem he was solving.
The operating issue
Strip away the novelty and Next was addressing four textbook revenue-management problems at once.
One: no-shows on inventory that cannot be resold. A high-end tasting-menu restaurant is the most exposed possible case. Capacity is tiny, each cover is worth a great deal, the meal is prepped to a count, and a party that does not arrive at 8:00 cannot be replaced at 8:20 because there is nobody waiting and the kitchen has already begun. Kokonas has spoken publicly and often about how large a problem no-shows were at conventional fine-dining restaurants and how nearly they disappeared under ticketing. Treat the specific magnitudes as his claims rather than as independently audited figures; treat the mechanism as sound, because it is arithmetic — a prepaid, non-refundable seat cannot no-show in any way that costs you money.
Two: demand that is wildly uneven across the week and the hour. Every restaurant has this. A famous one has it worse: Saturday at eight is oversubscribed by a factor most restaurants would find absurd, and Tuesday at nine-thirty is not. Ticketing let the restaurant price that difference openly rather than absorb it, moving some demand toward the hours it had.
Three: the working-capital problem this book calls cash timing. Chapter 33's whole subject is that a profitable restaurant closes when it runs out of money in a particular week. Selling dinner sixty days in advance inverts the normal restaurant cash cycle: you hold the guest's money before you buy the food, rather than paying the produce invoice on Friday for a dinner served on Tuesday. For a business with expensive, perishable inventory and a long prep cycle, that is a structural advantage, and it is available to almost nobody else in the industry.
Four: labor and purchasing precision. If you know exactly how many people are coming and exactly what they are eating — a fixed menu, a fixed count — you can buy to the cover and schedule to the cover. Chapter 13's purchasing and Chapter 19's staffing guide both become dramatically easier problems. Waste falls. The ideal-versus-actual variance narrows toward zero, which is a thing almost no restaurant achieves.
What it shows
Four things, and the third and fourth are the ones most people miss.
It shows that revenue management genuinely transfers — when the concept has fences built into it already. Next did not have to invent a fence. Its product was a fence: a single fixed menu at a fixed time, changed wholesale several times a year. A guest could not buy the cheap Wednesday slot and use it on Saturday, could not order à la carte, could not extend the meal. Every restriction that an ordinary restaurant has to construct artificially — and that its guests then resent — was already inherent in what the restaurant was.
It shows the size of the prize when you eliminate the no-show entirely. Look at it in this chapter's terms. A restaurant that prepays its seats has converted capacity utilization from a forecast into a fact. There is no refill rate to model, no waitlist to manage, no host improvising at 7:40. That is not a small operational gain; it is the removal of an entire category of variance.
It shows that the model's reach is narrow, and the reason is instructive. Ticketing spread to tasting-menu restaurants, chef's counters, wineries, and experience-led venues. It did not become the default for neighborhood American restaurants, brasseries, or anywhere a guest expects to decide what they want after they sit down. The reason is not conservatism. A ticket prices a defined product, and most restaurants do not sell a defined product. Bellwether's guest might spend \$31 or \$74 depending on whether they order a second cocktail and share a dessert, and the average of that distribution — \$46 — is a statistic, not an offer. You cannot ticket a distribution.
It shows the friction cost, honestly. Prepayment is a barrier, and barriers cost bookings. Alinea and Next could impose one because demand exceeded supply by a wide margin and because their guests were making a plan weeks out anyway. A restaurant with a soft Tuesday cannot. This is the single most important lesson to take away: the instrument was appropriate to that restaurant's demand curve, and applying it to a different demand curve inverts its effect. A prepayment requirement on a 44%-utilized Tuesday does not protect inventory; it destroys bookings you needed.
Outcome
Alinea and Next both operated for many years and both remained internationally prominent; the ticketing model became standard in the tasting-menu segment and is now offered as a feature by multiple reservation platforms rather than as a distinguishing choice. Tock grew into a substantial platform business, pivoted quickly to off-premise ordering during the 2020 shutdowns — a moment when the ability to take prepaid orders at a fixed time was suddenly relevant to the entire industry — and was acquired by a larger technology company.
The broader outcome is that prepayment, deposits, and card holds moved from exotic to ordinary across American restaurants between roughly 2011 and the mid-2020s. Very few restaurants sell full tickets. A great many now hold a card for large parties and peak windows, which is the diluted, low-friction version of the same idea — and is precisely what Bellwether's plan adopts in §24.7.
The lesson
A revenue-management instrument is not good or bad. It is matched or mismatched to a demand curve.
Ticketing is the strongest no-show control ever applied to a restaurant. It is also a booking barrier, a rigidity, and a statement to the guest that the restaurant's convenience comes first. In a room with three times more demand than seats, the first fact dominates and the rest are rounding. In a room at 44% capacity utilization on a Tuesday, the rest dominate and the first fact is irrelevant, because you had no no-shows worth preventing.
Bellwether's version of this lesson is the policy in §24.7: card on file for parties of six or more and for Friday and Saturday prime time; nothing at all everywhere else. That is the same instrument, dosed to the exposure. The exposure is concentrated in about four hours a week, and so is the policy.
The second lesson is quieter and it belongs to Chapter 33. Next's most underrated advantage was not that nobody no-showed. It was that the money arrived before the invoices did. If you ever find yourself with a legitimate way to be paid in advance for perishable capacity — a holiday prix fixe, a private event deposit, a gift-card program, a ticketed New Year's Eve — take it seriously as a cash instrument and not only as a revenue one. Chapter 33 will show you which week of your year that matters in.
Discussion questions
- Next's product was itself a fence. List three things about Bellwether's concept that make it impossible to ticket, and then describe the smallest piece of Bellwether's business that could be ticketed without breaking the concept.
- The chapter argues that prepayment is appropriate where demand exceeds supply and destructive where it does not. Bellwether's Friday and Saturday between 6:45 and 8:45 arguably meet the first condition. Should those four hours be ticketed? Argue both sides, then decide.
- Ticketing converts capacity utilization from a forecast into a fact. Which of the other two RevPASH levers — average check and dine time — does it also fix, and which does it leave alone?
- Kokonas's public claims about no-show rates are widely quoted and were never independently audited. As a reader of this book, what would you need to see before quoting one of those figures to a lender or a partner? What can you legitimately say without it?
- The cash-timing advantage of prepayment is real and mostly unavailable to ordinary restaurants. Name two other legitimate ways a restaurant like Bellwether could receive money before it incurs the cost of serving it, and identify the accounting and compliance question each one raises.
- Suppose a reservation platform offered Bellwether full ticketing at no charge tomorrow. What would you use it for, what would you refuse to use it for, and what would you measure after ninety days to find out whether you were right?