Editor’s Note: A version of this column was originally published on Franchise Times, a sibling publication of Food On Demand.
When delivery surged into the restaurant space, many brands struggled to get their high-margin drinks to homebound diners. The ubiquitous fountain drink lids failed and spilled soda all over drivers’ cars. The ice melted, the fizz faded and nobody got a straw when they wanted one.
Back then, some of the smartest operators I spoke with had simple guidance: Raise the price and deliver bottled soda. Ten-plus years later, does that advice hold true? In short, yes.
In a great discussion with Coca-Cola executive Josh Gurley, Franchise Times sibling publication Food On Demand explored why beverages still lag and offered ideas to boost beverage sales in digital channels.
Gurley spoke about ways to drive “beverage incidence.” (Dibs on that band name.) He sees potential in playing up personalization to all the special Gen Z butterflies who love the Coke Freestyle machines and taking advantage of the discovery behavior of ordering digitally versus a speed-focused drive-thru. Some of it works, and the soda supplier is working closely with third-party delivery companies to push it further.
Still, across digital channels, the percentage of orders that include beverages is about 14 percent lower for digital orders than offline. While that has gotten better in the last several years thanks to highly incentivized players like Coca-Cola, it’s still a distinct difference.
But China does it!
The most common refrain I hear is that China solved this years ago, and that’s fair. Meituan and Ele.me, two of the biggest delivery players, run a market that moves some 100 million food and drink orders on a normal day, with peaks that nears 250 million. The United States, by wild contrast, does maybe 14 million a day. A large share of the mix in China is beverages. Meituan says roughly 45 percent of orders were drinks, namely milk tea, coffee and the like.
But it works there for reasons that don’t port to a U.S. franchise P&L. The average Chinese delivery ticket runs about $7, couriers earn maybe 50 cents a drop, tipping isn’t customary, and the platforms and merchants, not the customer, absorb most of the delivery cost. Dense, electric-scooter fulfillment and heavy subsidies make a cheap drink worth moving. The sheer density of so many small-format, fulfillment-first locations is another major factor. Luckin Coffee passed Starbucks in China years ago and now runs north of 30,000 stores; Mixue, selling roughly $1 bubble teas and soft serve, operates close to 60,000 locations.
The U.S. market is almost the exact opposite when it comes to the economics of delivery. LendingTree pegs the all-in delivery premium at roughly 80 percent over pickup in the U.S., about $9.30 a ticket. Nobody pays a $9 surcharge to have one of even the fanciest dirty sodas with all the mix-ins carried to their door, so the standalone drink order barely exists. That’s the real beverage gap: Digital drinks don’t lag because customers aren’t thirsty. They lag because the drink used to ride along with a meal, and the meal is what justified the cost of the delivery.
But Starbucks does it!
The other glimmer of hope for drink delivery is the performance of Starbucks. And it is pretty incredible. In fiscal 2025 the company said U.S. delivery passed $1 billion in sales and grew nearly 30 percent year-over-year in the fourth quarter, with delivery available from essentially all eligible company-operated stores through DoorDash, Grubhub and Uber Eats. Starbucks says delivery orders run nearly twice the size of an in-store transaction and arrive in about 25 minutes.
With nearly 17,000 locations, Starbucks can sling coffee almost as quickly and efficiently as some Chinese brands with similar density per capita. But even Starbucks didn’t get here by shipping lone lattes; it also wins with food. Forty percent of orders include higher-margin food both by design of the ordering system and by the same incentive that drives people to order more when getting delivery: It’s really expensive, so you better get a muffin for later.
Also, people are physically addicted to coffee. The orders that don’t include food are very likely going to folks with no options between paying the price of delivery or paying the price with a caffeine-withdrawal headache.
The market matters
Strip away the headlines and the structural problem is simple math. A delivery order carries a very high fixed cost and a single drink is the worst possible thing to spread that cost across.
There’s also a substitution problem specific to beverages. For a lot of drink occasions, the convenience delivery sells is already covered by the drive-thru, c-store and by mobile-order-ahead pickup. The fastest-growing beverage chains have leaned into exactly that and skipped delivery on purpose.
Dutch Bros flat-out says it doesn’t deliver and it still posted a record $2.1 million average unit volume in 2025. 7 Brew hit $1.2 billion in system sales on a drive-thru-only footprint.
Operational issues abound
I love a crisp fountain drink, but let’s be honest: the cup is awful. Fountain drinks spill when you look at them wrong.
Sure, you can seal it for about the same per-drink cost, but then you need the sealing machine, you need to maintain that machine, and every drink needs to run through that machine. Hello bottlenecks.
The rest of the operational list is unglamorous and familiar. Carbonation fades and ice melts across the 25-plus minutes an order spends in a bag and a car. Double-wall cups, spill-proof carriers and tamper-evident seals help, but they all cost money and add steps in a space without a lot of upside.
There are some tricks. Dunkin’ integrated delivery into its point-of-sale system so orders fire close to when a driver is actually near, keeping hot drinks hot and iced drinks cold instead of letting them sit on a counter. But for most brands, that kind of tweak would just muck up the make line and create more grumpy delivery drivers.
Save yourself the headache. Just raise the price and toss a bottle in the bag.
Nicholas Upton has reported on retail and restaurant technology for more than a decade. His Tech Stack column aims to distill complex ideas into actionable insights. Send interesting tech topics to [email protected].
