How Apps Commandeered the Age-Old Idea of Takeout

Third-party delivery services have convinced us they are an essential part of our busy lives. But humans have managed to order food to-go for centuries.
Collage of images of Uber Eats delivery person ancient illustration of Roman thermopolia and old McDonalds
Photo-Illustration: Sam Whitney; Getty Images

On Tuesdays Smokey John’s BBQ delivers food to Irvine, Arlington, and Grand Prairie. On Wednesdays they drive to North Dallas, Carrollton, and Farmers Branch with East Texas–style brisket (more spice than the salt-and-pepper puritanism of central Texas) and ribs (saucy, owing to the proximity of the Carolinas and Memphis). The new hot-ticket queso is sold by the pint or quart with chips for dipping. It’s available only on Thursdays and Saturdays. So unless you live in Delivery Group C (Oak Cliff, Ce- dar Hill, Desoto, and Duncanville) or E (East Dallas, Garland, and Mesquite), you’ve got to come into the shop to get it.

When lockdown first hit Dallas, Smokey John’s was still getting pickup orders. But within two weeks, business had dropped 55 percent, and things were getting tight. Along with the restaurant, co-owners and brothers Brent and Juan Reaves had catering trucks just idling. So they added up their clientele, product, and wheels, and the Reaves brothers formulated a plan. On March 20, they went on Facebook Live to update their customers on the new delivery schedule, letting them know which areas of Dallas they’d cover on which days. They continued these broadcasts every night for months, maintaining the connection with customers that has long been a core part of the family business (originally named Big John’s by their father, John Reaves, until a fire in the late 1970s caused a customer to declare,“Y’all should call this place Smokey John’s instead of Big John’s!”).

A year before the pandemic, Smokey John’s had been offering delivery through the app Eat24, at a loss, just to expand the company’s reputation and get diners used to ordering online. Now customers in the five delivery zones around Dallas were regularly placing orders by five o’clock the day before, by phone, email, or fax. By the summer, revenue was up 15 percent from the previous year, without a penny of it going to a tech company.

Smokey John’s is the exception to the rule. It was one of the winners, the survivors if you will, of March 2020—because its owners conceived and executed a version of self-delivery. Pre-pandemic, restaurants were being flanked by app companies and left with two choices: They could refuse to participate and lose customers, or they could partner with the app companies and lose money on the ruinous commissions. Once Covid-19 forced an end to dine-in service in most places, restaurants were at the mercy of third-party delivery services, with little choice but to pay an average commission rate higher than the average restaurant profit margin.

Subsidized by the war chest of venture capital, these companies have in the past decade successfully gotten between restaurants and their hard-earned customers, aided by slick marketing that convinces us eaters we’re too busy to live without it and promises businesses they’ll grow sales while adapting to today’s uniquely fast-paced customer. There are many good reasons why getting carryout or delivery for dinner is a necessary expediency. But the idea that we have uniquely cultivated an existence that demands convenience to serve our mightily efficient lifestyles is more spin. We’ve always been busy. We’ve always craved convenience. No part of delivery is new, other than the predatory companies making it irresistibly easy—and using that ease to wedge themselves between restaurants and their customers.

Like us, the ancient Romans were busy. They had to make offerings to the gods and view chariot races and public executions with the rushed pace of our breakfast meetings, Pilates classes, and clarinet lessons. That’s why they invented fast food and takeout.

Thermopolia were businesses that sold food on the go, using long counters to store earthenware jars, called dolia, that kept food warm and enabled quick service. Think of the hot table at Chipotle—the one where cooked meats, rice, and beans sit in metal inserts, warmed by steam from below—minus the sneeze guard. Back then, not everyone had a kitchen in their home. At thermopolia, Romans could grab a quick bite of meat and cheese, spiced wine, lentils, fish, or nuts with a dash of garum, the liquid extract of fermented fish, similar to the fish sauce essential to Southeast Asian cooking, a condiment as ubiquitous to the ancient Roman as ketchup is to the modern American. These takeout spots weren’t rare. In the ruins of Pompeii, buried by the eruption of Mount Vesuvius in AD 79, more than 80 thermopolium counters have been discovered. Though the ancient Romans may not have had to pick up the ancient kids from ancient soccer practice, they were busy enough to have created fast food and takeout.

The much later invention of delivery is often credited to Italy. In 1889, according to legend, King Umberto and Queen Margherita asked chef Raffaele Esposito to bring a pizza to their palace in Naples. The element of royalty makes for a good fable about the proto-delivery. While the history of the thermopolium (along with tamales sold in open-air markets by the Aztecs of Central America) is backed up by archaeological discoveries, this origin story is suspect at best and, like many versions of history in which white people invent everything, probably apocryphal. True or not, around that same time in India, Mahadeo Havaji Bachche launched a Mumbai business shepherding hot meals between offices, homes, and restaurants. Bachche’s more formalized Dabbawala system is the clear progenitor of modern delivery. The tiffins, which are nested, cylindrical stainless-steel lunchboxes, are carried about India by train and bicycle with such dazzling efficiency and accuracy that the industry is admired and studied by business academics all over the world.

In America, some colonial-era restaurants offered carryout food to be picked up by servants. After the Civil War, an informal economy sprang up around train stops of Black women selling prepared food—one of the only entrepreneurial opportunities available, before or after emancipation. “For African American consumers, take-out was often less of a convenience than a necessity,” writes food historian Emelyn Rude, author of Tastes Like Chicken. “Blacks on a long journey or simply looking for a bite to eat away from home anywhere in the Jim Crow South were often forced to order their food as take-away in segregated restaurants if they wanted to eat at all.”

Until the middle of the 20th century, carryout in the US was mostly the domain of transit, of train stations and roadhouses. It wasn’t until after World War II, when new car sales quadrupled in America, that take-out and delivery exploded. With both the economy and the birthrate booming, Americans migrated from urban centers to newly developed suburbs. The GI Bill subsidized a massive expansion of postsecondary education and home ownership (often with zero down payment and low-interest loans), sometimes with preferred terms for new developments. The move to the suburbs and growth of car culture spurred the proliferation of carryout service, the specific popularity of pizza largely attributed to American GIs having served in Italy during the war and come home with a taste for Italian food. McDonald’s, created in 1943 and massively expanded in the postwar era, didn’t even add dine-in seating until 1963. For the first 20 years, it was all takeout.

The science and mechanics of takeout didn’t change much throughout this era. Made by the Bloomer Brothers (now Fold-Pak), the ubiquitous “Chinese takeout” container started life as packaging for oysters and scallops, popular for takeout in early-20th-century New York. In subsequent decades, various manufacturing developments allowed for the creation of paper, plastic, and Styrofoam containers that did better jobs of keeping food warm or cold, until the zenith of 1985’s McDLT, which came in packaging that kept the hot side hot and the cool side cool. But for about 40 years, not much else changed. To this day, we still don’t trust a plastic lid to stay on a container of hot soup—the business has never been terrifically tech-savvy.

These were the ways that, for a generation, we got restaurant food at home. Local restaurants printed takeout menus and slipped them under the doors of prospective customers. Most of us devoted a drawer in our kitchen to these menus, pulling one out for a break at the end of a particularly stressful week, or grabbed a familiar favorite from a drive-thru on the way home. That was the extent of it.

It wasn’t until the 1990s that technology began to fundamentally change this part of restaurants.

Recession-proof and perennially popular, pizza is inarguably the champion of American food. It is not just widely adored but endlessly adaptable, made dogmatically according to Neapolitan standards or adorned with butter chicken or pierogi, its crusts stuffed or composed of cauliflower, enjoyed in beautiful restaurants but also sold in the supermarket freezer aisle. Unlike one of its main competitors for our affection, the hamburger, it holds up perfectly when delivered. Made and sold high and low, it is a complete meal or a snack. Fittingly, pizza was the first physical product sold online. That inaugural digital sale, the Yuri Gagarin of ecommerce, was a large pepperoni with mushrooms and extra cheese from Pizza Hut, which launched PizzaNet in 1994. Though money changed hands only at the point of delivery, this was the antecedent of our contemporary one-click shopping experience.

In 2001 Papa John’s showed early tech savviness with its online ordering system, followed in 2010 by Domino’s, which became a trendsetter with its “pizza tracker” app, enabling the consumer to see at which stage of production or delivery their pizza is. In 2014 Domino’s introduced “Dom,” a voice-operated ordering feature that let you order by speaking, oddly replicating the telephone experience that technology was replacing.“I don’t even think that Domino’s is a food company anymore,” influential restaurateur David Chang told a Domino’s store manager on his television show Ugly Delicious. “I think of you as a tech company.” He intended it as a compliment.

Big chains like Red Robin, Famous Dave’s, and Panera Bread, with their economy of scale, resources to devote to digital development, and perceptive leadership that anticipated the growth in ecommerce sales, were able to place themselves ahead of the herd by developing self- delivery before the app-based tech revolution, when the wolves came hunting.

At the turn of this century, when Seamless was launched, mostly as a tool for offices to place large orders from restaurants and caterers, it didn’t register as a threat. Nor did Just Eat in Denmark (2001) or Grubhub (2004) or a host of others, which all began swallowing one another in a series of mergers and acquisitions that read like a tech version of biblical birth announcements: “And Just Eat acquired Hungryhouse from Delivery Hero, and Seamless merged with Grubhub, and Greylock Partners and Redpoint Ventures did invest in Just Eat, which begat SkipTheDishes.”

As with humans, the family of companies grew wider and more diverse. Here is a partial list of major competitors and also-ran companies in this sphere: Talabat, Snapfinger, Hungryhouse, Menulog, Eat24Hours, Ele.me, EatStreet, Eat Club, Munchery, Postmates, OrderAhead, DoorDash, ChowNow, Caviar, Foodpanda, Menu Group, SkipTheDishes, SpoonRocket, Deliveroo, Gopuff, Hello Curry, Foodora, Dunzo, Swiggy, Uber Eats, Wolt, TinyOwl, InnerChef, Maple, Tapingo, Rappi, Spring, Chowbus, and Glovo. As they proliferated and merged, these companies collected more detailed, more accurate customer data, the information aggregating into a tool that could anticipate and meet customer demands far more efficiently than even the most veteran restaurateur.

The arrival of the iPhone in 2007, followed by the 2008 recession and a whole generation of young engineers mobilized to create apps in a get-rich-quick land rush to be the next Facebook, was an indefensible assault on restaurants. A host with a reservation book and a landline was under-equipped to compete with order-placing technology that was suddenly in every diner’s pocket, feeding data into Silicon Valley app companies. Within a few years, these companies knew more about a restaurant’s customers—what we wanted, when we wanted it, how much we were willing to pay—than a small business ever could.

In 2016 a number of these companies made news by stopping their so-far-unobstructed growth. Before shutting down, Bento conceded that there was more money to be made in catering than on-demand delivery, SpoonRocket sold its technology to Brazilian food chain iFood, and Square tried to sell Caviar to Uber or Grubhub.

As word got out that third-party delivery was unprofitable, despite much-heralded sales, the conversation shifted. The problem wasn’t that the emperor had no clothes, that these companies—valued in the billions, with more investment cash pouring in every day—had hustled restaurants and investors. It was that of course delivering food wasn’t profitable. Not with human labor. When restaurant meals could arrive at our door via drones, robots, and self-driving cars, however, that’s when the sector would go from red to black. “If we don’t get the [autonomous car] software thing nailed, we’re not going to be around much longer,” Uber CEO Travis Kalanick told USA Today in 2016.

All of these enterprises prefer to be known as tech companies, as opposed to taxi or restaurant businesses. That’s true. They don’t deliver food. Many of them farm out the physical schlepping to other agencies, like Relay, Homer Logistics (acquired by Waitr), and Habitat Logistics. Bike and car couriers are never employees but “independent contractors,” granting the company the maximum exemptions from labor and employment laws regarding scheduling, overtime, sick pay, and wages.

Committed to the legal fiction that their product is something other than delivery, and that couriers are not employees, these companies skirt around the particulars of what service they actually provide, reminding you that you get food brought to you because of them, in some hard-to-quantify way. “Grubhub helps you find and order food from wherever you are.” “Uber Eats is the easy way to get the food you love delivered.” “Whatever you want, we get it. Order delivery for yourself or with friends and watch in real time as your Postmate brings you all the things you love.” It’s an impressive feat of copywriting, implying that they deliver food without stating it and therefore avoiding the liability of identifying themselves as delivery companies.

I would describe them in a different way. In my opinion, they are predatory enterprises that have figured out how to use technology to get between restaurants and their customers and then sell the customers back for a cut of the action. To me, that’s a scam. It isn’t just that some of these companies served the exact same product to roughly the same customers while taking the first 30 cents of every dollar. These apps don’t even make the process of making food cheaper. They don’t make the process of delivering food cheaper, either. They just enable an ease of sales. The tech companies extract value by charging the restaurant a commission, which can range from 10 to 40 percent, usually hovering around 25 to 30 percent. How can you take 30 percent off the top from a business with such thin margins? You can’t. “You’re not making a profit at that type of a haircut,” as one restaurateur put it to me. That’s a huge problem at a time when online order and delivery platforms, which barely existed until recently, constitute 10.89 percent of the $863 billion restaurant market.

Some restaurateurs charge higher prices for delivery orders to absorb the commission cost. Some third-party delivery companies won’t allow this. In early 2020, a group of New Yorkers sued Grubhub, DoorDash, Uber Eats, and Postmates, alleging a monopolistic practice that prevents competition, limits consumer choice, and forces restaurants into illegal contracts that effectively fix prices. The companies named in the class action declined to comment or did not respond to requests.

In the early days, a lot of restaurateurs looked at the commissions and turned their backs on the delivery apps, refusing to play ball. But then they found sales were dropping as these companies syphoned off customers. So they began to use the apps, many figuring that if their peers were doing it, there must be a way to make money.

The business aphorism “Don’t confuse revenue with profit” knows no political loyalty. A hospitality professor put this situation to me as an old management-school joke. “A manager tells the boss there’s good news and bad. The bad news is we’re losing money on every unit. The good news is that sales are up.” This joke doesn’t just describe the restaurants. It describes the delivery apps, too. Though companies like Uber, with their self-driving plans and diluted labor standards in the meantime, have multibillion-dollar valuations, they’re still operating at a loss. In 2018 Uber lost $1.8 billion. Before its initial public offering (IPO) in 2019, the company was rumored to be seeking a valuation in excess of $100 billion. With analysts decrying that as grossly overpriced, Uber lowered it to $82 billion and still wildly underperformed. DoorDash grossed nearly $1 billion in 2019 and still posted losses of $450 million. Despite that, the company pulled in another $400 million from investors. It closed 2020 by tripling revenue and doubling losses.

Big restaurant brands are able to profit from sales made through these apps because the demand for their product enables them to negotiate better rates. Grubhub’s (which also lost $155 million in 2020 despite increasing sales 29 percent to $1.8 billion) first-quarter 2020 results show that the company’s average profit from orders placed with independent restaurants was $4. For orders placed with “a partnered national enterprise brand,” it was $0. The apps need major brands but don’t make any money from them. So it’s the independents that pay the cost, your local ramen shop subsidizing a delivery service for McDonald’s. Until there are only a couple of players left in the field, making it possible for two competitors to increase fees, the business model of the tech-delivery industry makes no sense.

The only thing certain about third-party delivery is that we cannot predict the next steps from these companies. Not five years from now or even one. A decade ago they didn’t exist. Now they are more than 10.89 percent of the restaurant market and growing.

But it’s a corner of the restaurant industry where we can take simple, direct action through our choices. We can stop using these apps. Delete them from our phones. I just did it. Before we could use apps to order food, did we all starve to death? True, it means that I can no longer choose dinner from 200 different restaurants. I don’t think I ever did. Most of us order from the same half dozen places on repeat. If I really want food from a particular restaurant, I’ll call it in and pick it up.

If I find a restaurant near me that’s doing self-delivery, whether as a disaster adaptation like Smokey John’s or because they’re a pizza or Chinese restaurant that has always provided their own delivery, I’m going to support the hell out of them. When I’m going to indulge by eating something rich and delicious without cooking, I want the restaurant I’m ordering from to be making money.

If you do find yourself ordering from an app, tip in cash. You don’t want to Google the company later to find it’s one of the many that uses tips to drivers as a deduction from the promised hourly wage (like DoorDash, which stopped the practice in 2019 following public criticism). So at the very least, make sure the courier gets a solid tip and that it gets directly into their hands.

When I began writing and researching The Next Supper, the restaurant industry problems I was focused on—exploitation of workers, a fragile supply chain, the inequity of tipping, the hustle of third-party delivery—were so “inside baseball” they seemed almost like conspiracy theories. But because of the pandemic, these issues have become part of a public conversation about how the restaurant industry needs to change. Deleting third-party delivery apps may be difficult for some of us, hooked as we are on the convenience or routine. But for diners who love restaurants and want to support them, it’s one of the simplest, most effective things you can do.


Adapted from The Next Supper: The End of Restaurants as We Knew Them, and What Comes After by Corey Mintz, copyright © 2021, PublicAffairs, an imprint of Hachette Book Group, Inc.


More Great WIRED Stories