Chip City began in Astoria, Queens in July 2017, out of a competition between two childhood friends, Peter Phillips and Teddy Gailas, over who could make the better cookie. It sold thick, fresh-baked cookies on a rotating menu, built a genuinely loyal following, and attracted serious money.
Danny Meyer’s Enlightened Hospitality Investments put in 10 million dollars in 2022 and a further 7.5 million in a Series B in May 2024, taking the total to 17.5 million. At that point the chain had 38 locations. The stated plan was to grow past 50 nationwide, and it did: at its peak the brand had more than 45 locations across eight states, including New York, New Jersey, Connecticut, Florida, Illinois, Maryland, Massachusetts and Virginia.
By October 2026 the website listed 24 locations, across New York, New Jersey and Texas. The three Virginia sites closed, including one at Tysons Corner Center that had opened in late August 2025 and lasted about a year. Connecticut and Massachusetts appear to have gone entirely. A rival cookie shop in Arlington offered to honour Chip City gift cards and invited its former employees to apply for jobs.
The product was not the problem. The product was the thing that got the company 17.5 million dollars and a partnership with one of the most respected operators in American hospitality.
What went wrong is a geometry problem, and it is the single most common way good small food businesses break when they take growth capital.
A cookie shop in Queens works because of density. Enough people walk past, enough of them come back, the staff can be managed from nearby, deliveries are short, and the founders can be in the room when something goes wrong. Everything that makes the unit economics work is a function of being close to other units and close to the people running it.
Open a site in Virginia, one in Massachusetts, one in Illinois, and the model changes entirely. Each of those sites is alone. It has no neighbouring store to share staff with, no local reputation to inherit, no regional manager who can be there in twenty minutes, and a supply chain built for a different state. Fixed costs that were spread across a cluster now sit on one shop. The brand recognition that filled a queue in Queens does not exist in a mall in Tysons Corner, where the shop is simply a new cookie place next to the pretzels.
Growth capital accelerates whatever is already true. Where the model is dense and repeatable, it produces a chain. Where the model depends on proximity, it produces a map with pins too far apart to support each other.
The retreat to three states is not a failure of nerve. It is the company doing the arithmetic a second time and concentrating on the markets where the density exists. That is the right decision, and it is far more expensive after the fit-outs than before them.
The questions worth asking before the expansion money arrives are narrow and unglamorous. What exactly makes this site work? Which of those things travels? And how far apart can two shops be before they stop helping each other?
Chip City had the cookies and the capital.
What it needed was the second shop to be nearer the first.
