On the Clock: What Low-Wage Work Did to Me and How It Drives America Insane

by Emily Guendelsberger

For Rajiv, who is the best

I wish you well, children.

This book is an examination of the day-to-day experience of low-wage work in America in the mid-2010s. There was a lot of totally standard journalism and research involved, but because what interests me most is the experience, I worked three jobs undercover to be able to accurately describe that experience. I mean, kind of undercover. I didn’t volunteer that I was writing a book, but I also didn’t lie or misrepresent myself. Here’s how that worked, if you’re interested. About the timeline: The newspaper I worked at closed in October of 2015. I spent part of November and all of December that year in Southern Indiana, across the river from Louisville, Kentucky, working at Amazon. I spent most of the summer of 2016 in western North Carolina working at Convergys. I spent September and October of 2017 in San Francisco working at McDonald’s.

What does “in the weeds” mean to you? I’ve been asking people that for a couple of years now—it’s become a sort of hobby. There’s two definitions, and you can often tell a lot by which one a person knows. First, there’s what I call the academic definition: “To be bogged down in the minute or unimportant details of a large project.” I heard this a lot in the ten years I spent working in newspapers. Then there’s the waitress definition: “To be harried or frantic because there’s more work on your plate than you can do at a reasonable pace.” A key part of this definition is a feeling of desperation and hopelessness—being unable to catch up, even though you’re working as fast as you possibly can.

Dad sighed, and gave me some fatherly advice that stuck with me for a long time. “Well, life isn’t always fair, for one. But there’s dignity in working hard and doing your job well,” he said. Yes, even if that job was scooping ice cream. If you can take pride in doing your work well, that’s the key to a happy life, no matter what your job is. Be the best at whatever you do, try harder than everyone else, make yourself indispensable, and you’ll succeed. It’s the American Dream. Chastened, I resolved to be the best ice cream scooper I could possibly be. For the rest of the summer, I tried to be that overeager, kiss-ass robot. I tried to make my cones beautiful. I made myself smile and be extra friendly when I was exhausted. I cleaned equipment before being asked to. I even asked if there was anything I could mop. I kept hoping that if I worked hard enough, the dignity my dad talked about would just materialize, like a runner’s high. But it’s hard to feel dignified when your right arm is never not sticky or when a jackass crew of boys runs you ragged while pretending not to recognize you from history class. Anyway, my manager didn’t seem to notice my attempts to be the best damn ice cream scooper in the land. All us counter kids were summarily fired at the end of the season, and I moved on to a string of service jobs that took me through high school into college and my early twenties—dishwasher, cashier, receptionist, mall photographer, church pianist, grocery stocker, projec...

Even now, real, respected people with real, respectable jobs express confusion about things that make me wonder how long it’s been since they’ve had a conversation with anyone making less than $100,000 a year. These vexing puzzles include: • Why do American employers complain they can’t find good workers to fill open positions? • Why has the life expectancy of middle-aged white Americans fallen off a cliff in the past decade? • Why is the country ready to riot over jobs—immigrants taking them, trade deals killing them, Wall Street destroying them? • Why are depression and anxiety so widespread when this is one of the best times to be alive in history? • Why do people vote against their self-interest? • If the data says everything’s so great, why is America freaking the fuck out? You can make a lot of money explaining away the gap between data and reality in ways that flatter puzzled wealthy people. But if you’ve had a service job in the past decade, I’ll bet that some of the answers are probably as obvious to you as why millennials aren’t buying yachts. I’ll spend the next few hundred pages trying to make it just as obvious to all you readers, but the short answer? The bottom half of America’s labor market lives in the weeds. All the time.

A modern McDonald’s job, for example, isn’t the leisurely activity implied by “flipping burgers.” Fast food today is intense. In McDonald’s phone-book-size operations and training manual, every task has a target time in seconds, as in the assembly of a burger: Target order display screen reaction time: 5 seconds. Target toast time: 23 seconds. Target sandwich assembly time: 22 seconds. Target sandwich wrapping time: 14 seconds. Target order assembly time: 16 seconds. “Guests should wait no longer than 60 seconds from the time the order is being totaled until the order is presented,” the manual advises, and the total time between a guest approaching the counter and receiving her food should never be more than three and a half minutes. Until pretty recently, it was much too difficult to track and enforce guidelines this specific. Today, though, monitoring equipment integrated into the tools workers use to do their jobs can clock and track nearly any task a worker does in real time. And the street runs both ways—those same systems can be set to harass, nag, startle, or otherwise trigger a worker’s stress response every time she lags behind the desired pace. (The next time you’re in a busy McDonald’s, for example, note the maddening, near-constant beeping of alarms.)

The other big advance that’s made life miserable for low-wage workers is algorithmic scheduling. Work schedules that used to be drawn up by managers now rely heavily on algorithms that analyze historical data to predict exactly how much business a store can expect in the upcoming week. As it’s most accurate with the most recent data, this means many workers’ schedules vary wildly week to week and are made and posted the day before they start—making it impossible to plan anything more than a week in advance. Businesses also save a ton of money by scheduling the absolute minimum number of workers to handle the predicted business. And they save even more by scheduling slightly fewer people than can handle the predicted work at a reasonable pace. If workers can push themselves to cover the duties of a sick coworker, doesn’t that just mean they’re not giving it 100 percent the rest of the time? Why can’t they work that efficiently every shift? The answer’s obvious if you’ve covered for a sick coworker at a fast-paced job—because you’re stuck in the weeds the entire day, and just because you can put up with a miserable day once in a while doesn’t mean that the weeds are a sustainable place to live. From a boss’s point of view, though, the weeds are where workers should be—at maximum productivity, all day, every day.

That’s why I decided to go experience this brave new world myself—so readers can get an idea of what the modern experience of low-wage work feels like. I went to work at an Amazon warehouse, a call center, and a McDonald’s—three places that are fairly representative of the future of work in America. I spent between one and two months at each, and I worked them for real, with every ounce of my Dad-infused work ethic. And at each one, productivity-enforcing technology constantly corralled me and my coworkers into the weeds like a sheepdog snapping at a herd’s heels. Working in an Amazon warehouse outside Louisville, Kentucky, I walked up to sixteen miles a day to keep up with the rate at which I was supposed to pick orders. A GPS-enabled scanner tracked my movements and constantly informed me how many seconds I had left to complete my task. Working at a call center in western North Carolina, I was lectured about how using the bathroom too often is the same thing as stealing from the company, and had the minutes I spent in the bathroom tracked in a daily report sent to my supervisor. Working at a McDonald’s in downtown San Francisco, we were underscheduled to the point of a constant, never-ending line of customers—everyone worked at the frantic speed of those in-the-weeds waitresses of my youth all shift, nearly every shift. I did plenty of research beforehand, and I’d heard crazy things about how stressful each job would be—each in its own special way, like Tolstoy’s unhappy...

Amazon.com went live on July 16, 1995, selling books out of a two-hundred-square-foot basement room in an industrial area of Seattle that had once been used for band practice and still had SONIC JUNGLE spraypainted on the door. By 1997, as founder Jeff Bezos was trying to find funders for Amazon’s IPO, the warehouse had relocated to a 93,000-square-foot space on Seattle’s Dawson Street. Still, investors were skeptical about Bezos’s business plan. “If you’re successful, you’re going to need a warehouse the size of the Library of Congress,” Bezos remembers one man explaining before turning him down. The modern Library of Congress, the largest library in the world, contains 2.1 million square feet, according to the National Park Service. The fulfillment center I’m standing in alone contains more than 2.5 million square feet of storage space, and, as of 2019, when this book will first be published, it won’t even be in the top ten biggest of Amazon’s 175-odd fulfillment and sortation centers in the US.

SDF8 doesn’t look like much from the outside. Stuck in the enormous shift-change traffic jam this morning, I could just make out the corner of a big beige warehouse in the predawn darkness, looking no different from the many, many big beige warehouses around here. The southern tip of Indiana, right across the Ohio River from Louisville, Kentucky, is a logistics hub—that’s an area with a cluster of warehouses at some crossroads of air, truck, train, and river shipping routes. “Locating in Louisville gives your company the ability to move your products to 80% of the world’s population in less than 48 hours” is the pitch from the city’s economic development group. FedEx, UPS, and the postal service have big facilities here; so do eBay, Caterpillar, Wayfair, and many, many others. The ideal logistics hub also has lots of cheap space, offers business tax breaks, and is home to a lot of underemployed people who used to work in manufacturing.*

The train of newbies keeps walking, now passing an area where hundreds and hundreds of those yellow bins are stacked in shoulder-high towers. Each bin is roughly big enough to hold a small woman with her knees drawn up to her chest, and I immediately, desperately want to ride one around the conveyor belts like a roller coaster. Later, when I get out into the mod for real, I’ll note that every potential boarding point for a conveyor-belt joyride has several dire signs warning that it will result in the joyrider’s gruesome death, so I guess I’m not the only one.* And speaking of dumb ideas, I’m now walking while staring straight up at the conveyor belts, so I walk directly into the man in front of me as we bottleneck at a door. “SORRY! SO SORRY!” I yell-apologize, trying to speak loud enough to be heard over the din. SDF8 isn’t painfully loud, but it’s almost impossible to hear anything else, like a white-noise machine cranked up to 11. The guy, a tall black man in his early twenties, turns and flashes a grin with a gaping hole where his front teeth used to be. “IT’TH OKAY, I’M THTILL ATHLEEP, TOO,” he shouts back as we filter into a room already packed with more new temps.

The 2012 first-person account in Mother Jones2 detailed the week reporter Mac McClelland spent working in a warehouse for “Amalgamated Product Giant Shipping Worldwide Inc.,” generally understood to be lawsuit-avoidant shorthand for Amazon: The culture is intense, an Amalgamated higher-up acknowledges at the beginning of our training.… We don’t want to be so intense, the higher-up says. But our customers demand it.… The gal conducting our training reminds us again that we cannot miss any days our first week. There are NO exceptions to this policy. She says to take Brian, for example, who’s here with us in training today. Brian already went through this training, but then during his first week his lady had a baby, so he missed a day and he had to be fired.… Okay? Everybody turn around and look at Brian. Welcome back, Brian. Don’t end up like Brian.

Workers also tended to roll their eyes over the response to the Pennsylvania heatstroke story—specifically, to the idea that this was unusually heinous. Even the highest temperatures mentioned in the Morning Call sounded pretty standard (though unpleasant) to people who’d worked in other warehouses. High temperatures in the summer are normal, because it’s absurdly impractical to air-condition enormous, unshaded, uninsulated warehouses with big open truck bays. That’s kind of a moot point now, anyway—after the bad publicity, Amazon spent $52 million retrofitting most of its US fulfillment centers with air-conditioning. According to one worker, this is going about as well as the laws of thermodynamics would lead you to expect: Amazon installed huge amounts of A/C into all of its warehouses and it took the temperature down, like, two degrees in the summer. Massive waste of money, but they did it anyway to try and avoid headlines like these. Public opinion, even if that opinion is grossly ignorant and misinformed, is more important than money.

Workers said the Mother Jones piece got closer to nailing the “existential shithole” aspect, but was still a little misleading—the reporter wasn’t used to warehouse work and hadn’t stayed anywhere near the two weeks it takes most people to adjust to the workload. Of course her experience was agonizing—everyone’s first two weeks are agonizing.

The real problem, I was told, was the subtler plague of injuries from working through pain for fear of getting fired for low productivity numbers. That’s where Amazon really stood out—repetitive stress injuries. “Yes. Yes. Yes. There are lots of repetitive stress injuries,” said a man in his early thirties who’d been working as a picker in a Tennessee fulfillment center for three years. “I know people who developed various joint problems working there, and I now sometimes experience discomfort in my knees, which was never the case before.” “I saw plenty of older women who literally could not do their jobs any longer; their knees were not holding up. Mandatory eleven-hour shifts at a ridiculous pace will do that to you,” wrote another former worker. Many others mentioned injuries that lingered long after they left Amazon—particularly a problem for temps, who aren’t eligible for workers’ compensation.

When you clock in at Amazon, the first thing you do is grab a scan gun—a bit like a scanner you’d see at the grocery, but with an LCD screen that tells you what task to do next and starts counting down the seconds you have left to do it. It also tracks your location by GPS—and you take it everywhere with you, even the bathroom. Failure to stay ahead of the countdown—to make rate—was grounds for termination, regardless of why. “Anyone not getting the numbers they want to see, they get rid of them,” wrote one worker. “And by get rid of them, I mean deactivate their cards, so they turn up for work and can’t enter the building.” “You’re at a dead sprint your entire shift,” wrote another picker, one of many who said their rates were often set unrealistically high, which forced them to spend entire ten-and eleven-hour shifts in the weeds. “I was a young, average-shaped male, and while I could do it, it was not easy.” “Only one or two employees a day, out of forty, makes rate in the department I work in,” wrote another, saying managers sometimes would “revoke talking privileges until people start making rate.” “The job is hard on your body, the company pays as little as it can get away with, and the raise schedule for hourly workers is the worst of any company I have worked for, including big, bad Walmart,” wrote another. “Amazon doesn’t really treat its workers any better than Walmart, yet largely escapes the negative publicity.”

Simon Head’s 2013 book, Mindless, on the rise of productivity-enforcing technologies, spends a whole chapter just on Amazon and Walmart. “When I first did research on Walmart’s workplace practices in the early 2000s,” Head writes, “I came away convinced that Walmart was the most egregiously ruthless corporation in America.” In its analysis of the growth of US labor productivity between 1995 and 2000… the McKinsey Global Institute found that just over half that growth took place in two sectors, wholesale and retail, where Walmart “caused the bulk of the productivity acceleration through ongoing managerial innovation that increased competition intensity and drove the diffusion of best practice.”

When I finally get to a bathroom, I’m pleasantly surprised to find that it’s big, and even clean. Like, Canadian rest stop clean. I dash into an open stall, relieved. A flyer on the back of the door informs me there will be free pizza at lunch today. Wow, that’s pretty cool of them, I think, surprised at how much… nicer everything’s been than I’d expected. Then, as I reach for the toilet paper, I notice three messages scratched into the metal of the wall-mounted feminine-product disposal bin: I HATE THIS PLACE ME TOO THIS PLACE SUCKS DONKEY BALLS I will think of this little found poem many, many more times before Christmas.

Finally I get to which job I’d prefer, though I’m almost certainly going to be put in picking or packing regardless of how I answer. Packing is standing in place for ten hours, grabbing orders off a conveyor belt, and sealing them into shipping boxes; picking is walking around finding the items people have ordered and sending them off on a conveyor belt to be packed. Picking is generally regarded as the least desirable job at Amazon, with packing a close second. My busted right knee can tolerate lots of walking much better than lots of standing. Plus, as I understand it, packing stations are offset so you can’t talk to each other, whereas I assume you sometimes run into other pickers out in the shelving system. I select picking as my first choice. This mildly surprises the woman I’m eventually called up to speak with half an hour later. She shrugs as if to say “Your funeral,” and asks whether I’m looking for day or night shift—day is 6:30 a.m. to 6:00 p.m.; night is 6:30 p.m. to 6:00 a.m. “Night pays a dollar extra,” she adds, but I opt for day’s $10.50 an hour. She tells me, somewhat apologetically, that even a volunteer picker who can start tomorrow isn’t going to get a schedule with two consecutive days off right away. She hands me a list of potential schedules, out of which having Tuesdays and Thursdays off seems like the least bad option. And, to my surprise, that’s it. My interviewer types as I suck on an oral drug test, then puts it in a biohazard bag and congratula...

“We will not consider excuses for being late outside of approved exceptions,” says Miguel. Temps don’t have sick days or vacation: instead, we have a point system that Miguel explains was developed to give us as much freedom with our schedules as possible. “You have six points: if you’re at six points, your assignment with Amazon will end,” says Miguel. “Try to keep your points low—that way you will have flexibility in case of an emergency.” Miguel clicks forward to a slide listing the many ways we can rack up points. “Say I’m late to my shift or late returning from lunch less than an hour—how many points do I get?” asks Miguel. Half, we chorus, reading the answer off the screen verbatim. “If it’s more than an hour?” One. “Now, if I leave work up to an hour before the end of my shift?” Half. “And if it’s over an hour?” Whole. “What if I don’t come to work?” Point and a half. “What if I don’t come to work and also don’t call out?” Three. “If you do not finish your shift and walk off the job without telling anybody?” Termination.

We will clock in and out by scanning our white badges four times a day. “Lunch is gonna be a thirty-minute break, unpaid, and you will also get two fifteen-minute breaks, paid. “Now,” Miguel asks us, “what is it that you are not going to like about the breaks?” A former worker raises his hand. “Your break doesn’t start once you get out the door. It’s your last item you scan—that’s where your fifteen minutes starts. And it takes you ten minutes to get out the door, so you’ve got five minutes to get back and make your first scan.” Miguel nods at this correct answer. “Breaks are measured from last scan to first scan. So if your break starts at ten, then ten is when you make your last scan. And then by 10:12, or 10:13 at the latest, you head back in to your work area.… You know,” he pauses, “the easiest thing in the world is taking too long a break.” The former worker pipes up again. “You’re gonna get tired and lose track of time.” Miguel nods. “You will be coming back at 10:19, 10:20. If you do that, then—” “The coach is going to be there. Guaranteed. Guaranteed,” the man again interrupts. “It’s more than just guaranteed,” says Miguel. “They’ll want to know the reason you took those extra few minutes. If this happens for a reason, then you will be fine! But if you don’t have a reason, then you will not be fine. You are going to be fired.” Miguel seems pleased with this odd turn of phrase, as if he came up with it himself.

Say you worked a full forty-hour week, but Amazon only paid you for thirty. Wouldn’t they be stealing from you? Wouldn’t you be mad? Yes! we chorus. So then doesn’t Jeff Bezos have a right to be mad if you’re clocked in for forty hours but only actually working for thirty? Aren’t you stealing from him? Yes, we agree, with less enthusiasm. And, unlike other places we may have worked, Amazon will know if you’re only on task for thirty hours. “Amazon knows how your day is spent,” Miguel says. The company can track every second of a worker’s productivity throughout the day. “We have a saying in Spanish—Siempre va a te ver. That means There’s always an eye out there on you.” The “eye” isn’t cameras—it’s the scanner guns former workers talked about, which wirelessly upload your location and how long it’s been since your last bar-code scan in real time. If you’re going too slow, the scanner will display a message telling you so. If you go too slow too often, a “coach” will come find you to tell you in person.

After lunch, we learn how to navigate the mod. The first thing you’ll do every morning, Michelle says, is grab a scanner gun. She holds up her own. It looks just as it’d been described to me—like a cordless, handheld scanner in a grocery store, but with more buttons and a square LCD screen. Michelle scans the blue badge hanging from a lanyard around her neck—BEEP—then demonstrates how to push a long sequence of buttons in order to request work. She tells us not to worry if we don’t remember it all at first—the sequence is printed on a laminated card in our orientation packets that we can hang off our lanyards. Michelle’s scanner now reads SCAN A NEW TOTE—tote, we learn, is the Amazonian term for one of those omnipresent yellow bins—above an intimidating-looking alphanumeric code: P-3 A198 D221 This string is the coordinates of the one item among millions we’re about to go find, Michelle says. It may look crazy at first, but it’s easy if you pretend you’re finding someone’s apartment. That letter A tells you you’ll be in East Mod—that’s like a state. P-3 means you’ll be on the third floor of the mod—the floors are like different cities. The two three-digit numbers give the apartment building’s address—221 198th Street. The D means you take the apartment building’s elevator up to the fourth floor—you open the fourth drawer from the bottom. Then the item’s barcode is like the apartment number.

“So what do you do to keep yourself from going nuts out here?” I ask. Michelle laughs, and so does Jasmine. They admit to singing to themselves a lot—plus you do sometimes run into other pickers out in the mod. “Every once in a while I’ll make conversation with somebody about something,” says Jasmine, “because people, they’ll talk to you about anything.” Michelle nods knowingly. “Like, it gets bottled up?” I say. “Yeah!” says Jasmine. “It’s almost like jail—like, after you’ve been inside so long—” “And you can’t see the outside,” Michelle finishes. “Unless you find a window, which is, like, impossible.” We’re now pretty deep into the mod, surrounded by thousands of pieces of modular, double-sided, floor-to-ceiling cardboard shelving. It’s incredibly visually busy—nearly every surface has a color-coded sign on it. I can’t see either end of the corridor we’re on. I know if I just walked straight in any direction long enough I’d eventually get to an edge, but it still feels creepy, like you could get lost in here forever.

We eventually bumble our way to the correct coordinates, count up from the bottom, scan the bar code on the proper drawer—BEEP—and pull it open. Inside is a bizarre selection of twelve or fifteen articles of clothing; no two are the same. Michelle explains that a while back Amazon figured out that “chaotic storage” made the most sense—if a picker needs to get a pair of medium black socks, it’s much faster to find black socks among a bunch of dissimilar items than to find a medium in a drawer full of black socks of all different sizes. “Now”—Michelle shows us her scanner again—“here’s the description, at the bottom, of what you’re picking.” The item we’re looking for is an Ugly Christmas Sweater, medium, green. It’s obvious where it is, but Michelle picks up and scans a hat instead—DEE-DOO-DEE-DOO. The tritone beeps clearly mean “Nope!” The scanner directs Michelle to scan the drawer again—BEEP. This time she picks up the bagged sweater, scans its bar code—BEEP—and tosses it into the yellow tote. I’m a little charmed to see that the LCD screen says THANK YOU! before instructing Michelle to scan her tote again—BEEP—and giving her the coordinates of the next item. That’s pretty much all there is to picking. We take turns finding items under Michelle’s supervision for the rest of the day. When the yellow tote gets filled to the two-thirds mark, we push the cart over to the nearest conveyor belt, close the tote out on the scanner—BEEP—send it gliding off to parts unknown, then ...

The next day, my alarm goes off at 3:45 a.m., and this time I do arrive early. I kill some time investigating the weird vending machine from the previous day. It’s full of single-serving packets of generic over-the-counter medicine—mainly mild painkillers like Tylenol and Advil, though there’s antacids, too. After standup meeting, my group reassembles around Michelle, minus tiny blond Teena. Maybe she decided this wasn’t for her; maybe she failed her drug test; maybe she overslept and got fired. People disappearing without explanation will be another constant theme of my time here. The second day of training is the same as the first. Honestly, one day would have been enough. The system just doesn’t leave many possibilities for screwing up. As Michelle says, if you don’t know what to do, just check your scanner—it’ll tell you. We work on trying to get fast enough to beat the clock, which manifests itself visually as a blue bar at the bottom of the LCD screen that gets shorter and shorter as your remaining seconds tick by. By the end of the day, I can beat the clock most of the time.

All day, I’m charmed that my scanner thanks me every time I drop a yellow tote off at the conveyor belt. Watching it sail off into the distance gives me a weird feeling of satisfaction, like I really am helping customers fulfill their dreams. You’re welcome! I think.

Counterintuitively, Zappos used to offer its workers a lump sum to quit, even in the warehouses. The amount of money started at $2,000 and increased the longer you’d been there. The theory was that this would weed out people who didn’t really want to be there and that it would keep the culture positive. It apparently worked—Katie turned the money down twice, and so did most people. CEO Tony Hsieh later wrote in Delivering Happiness, a history of Zappos, that less than 1 percent of Zappos employees took the offer. This is why, Hsieh writes, he used to let journalists roam free around Zappos buildings. “Unlike most companies, we don’t give reporters a small list of people they’re allowed to talk to. Instead, we encourage them to wander around and talk to whoever they want. It’s our way of being as transparent as possible, which is part of our culture.” Katie quit not too long after Amazon acquired Zappos and began changing its warehouses into fulfillment centers. “It was just… different. It wasn’t fun anymore. It was go, go, go all the time, like we were just robots to them.”

I start to respond but get ambushed by a yawn midway through the sentence. It’s barely eight. When did I get so tired? It didn’t feel like I was doing that much, and my step counter said I’d only walked six miles. But I can’t keep my eyes open. I barely get my shoes off and into bed before I’m asleep—teeth unbrushed, face unwashed, fully clothed. When my alarm goes off for my second full day of picking, it is the worst my body’s felt in my entire life. And I’m not unfamiliar with pain. As a kid, I went through years of complicated corrective surgeries on my legs, followed by some nasty physical therapy. I’ve broken bones. I get incapacitating migraines. Playing rugby in college, I was once matched up against a woman named Rhonda the Tank. This is worse. It feels like I’ve been hit by a garbage truck. Everything hurts. My feet are the worst, but my back, shoulders, arms, and neck feel terrible, too. My hips, knees, and thighs ache from all the squats to pick from floor-level drawers. My right wrist, hand, and fingers ache from operating the scanner gun. My right elbow aches from pulling open hundreds of drawers. I even have a throbbing headache. I groan involuntarily as I force myself to swing my legs over the side of the bed. The carpet feels like it’s made of knives when I put weight on my feet, but I force myself to limp upstairs to the shower. Even before I’ve made coffee, I swallow a double dose of Advil and toss the bottle in my purse for later. I will need it. At SDF...

Finally, I do a full squat to retrieve an item from a bottom drawer and my body mutinies. Stand up, I order my legs for the hundredth time today, but it’s as if they’ve gotten fed up with all the abuse and hung up on my brain. Stand up, you idiot, my brain screams as I slowly topple backwards into a sitting position, but it’s just not happening. I sigh. Might as well rub my feet as long as I’m down here. I start to take off my shoes and am slightly horrified to find that my feet are so swollen they’re straining my shoelaces. Untying the laces feels sinfully good. The “sinful” part is reinforced by my scanner, which has started notifying me that I have zero seconds remaining to pick my next item. Fuck you, I think, reaching above my head to drop the horrible thing on my cart, out of sight. I try to rub some life into my horrible-looking feet. How many minutes of this can I get away with before some Time Off Task algorithm alerts a manager? If you’re deep in the mod, it can take fifteen minutes to get to a bathroom and back—if I keep it under fifteen minutes, maybe it’ll just look like a bathroom break. I desperately want to believe this, so I do—in retrospect, I realize that the scan gun’s GPS would have given me away.

It’s pretty clear that whatever algorithm plots our pick paths around the warehouse is brilliantly engineered, immensely complicated, and set to keep people from getting within speaking distance of one another. The mod, though it’s comparatively swarming with pickers during peak, is a very lonely place. I catch glimpses of people pushing carts or sorting through bins off in the distance, but I rarely get close enough to another human long enough to say hello, much less chat. Keeping us isolated makes logistical sense because the alleys between shelves are so narrow. But it also eliminates the opportunity for inefficient workplace chatter, which I’m convinced is a goal rather than a side effect. When I’m sent over to learn packing one day, the workstations are set up catty-corner from one another, making it impossible to talk.

Then I remember the painkiller vending machines. Supposedly they’re free with a swipe of your ID badge, which is good, because we’re not allowed to bring our wallets inside. I’ve seen other vending machines around the mod, but the only one I’m positive I can find is back by East Mod’s standup area, a five-minute walk from where I’m sitting. Get up get up get up, I yell at myself. You cannot lose this job! I drag myself to my feet, wincing, and limp toward where I hope there’s a staircase to the ground floor. But when I finally arrive at the vending machine, it doesn’t recognize my badge. I thunk my forehead against the glass in despair, staring in at the rows of little foil packets. So close. So far. A woman notices me being pathetic and comes over. “Let me guess—it’s your first week,” she says, pity in her Kentucky drawl. She’s white and middle-aged, with the blue badge of a full-time Amazonian and the air of lower management. I tell her it’s my third day out of training, and she grimaces. “Yeah, well, the second day’s worse than the first and the third’s worse than the second. But I promise this is as bad as it gets. If you can make it through the first two weeks, it gets easier. It really does.” She looks genuinely sympathetic. “Is the machine not working with your badge? It should…” She takes my ID and tries swiping it herself, with the same result. She frowns. As she tries a few more times at various speeds, I ask about the vending machines, which are apparently new. ...

I have it so much easier than nearly all my coworkers. At least half the East Mod pickers look past retirement age. Most younger people have kids, and they talk on smoke breaks about doing chores and making dinner after their shifts. There’s even a few very pregnant women picking, which flabbergasts me every time I see one out in the mod. I thought my ridiculous-looking warehouse shoes would help me blend in, but everybody else wears sneakers that are cheap, old, or both. I see an alarming number of Chuck Taylors, which would give me shin splints in thirty seconds. Despite all my research, I’m embarrassingly unprepared for what “normal” means outside the white-collar world, and I’ve grossly misjudged what $10.50 an hour is worth to a lot of people. I’d expected the pain. I’d expected the monotony. I hadn’t expected so many people to regard this as a decent job. Because Amazon does pay better than comparable jobs in the area. And I’d known, academically, that 80 percent of low-wage workers in the private sector don’t get any paid time off. I hadn’t really understood it, though. Unpaid time off still sounded like a laughable benefit to me, but tons of my coworkers were grateful for it, saying it really did give them more flexibility than most other jobs. White and blue badges alike say they’d take the worst job here over the best job in fast food, a sector discussed with a universal shudder. And if you hustle enough to get that blue badge, the health insurance is good, and t...

As I inch out of the massive parking lot, I spot Darryl in the large huddle of people at the bus stop. It’s freezing and miserable, and we’re going the same way. I consider stopping to give him a ride. No. Fuck Darryl. I don’t care. I don’t want to make small talk. I don’t want to delay my bedtime for however many minutes I’ll have to go out of my way. I don’t want Darryl to get the wrong idea, and I don’t want to feel obligated to give him rides in the future. I just don’t care. And I want McDonald’s, right fucking now. So I pretend not to see him. I find it hard to explain the needling shame I still feel about this. It’s not a big deal, and I’m positive Darryl wouldn’t hold it against me. But I like to think of myself as someone who’d offer a nice kid a ride home after a truly shitty day of work. Right now, though, exhaustion has shrunk my circle of empathy to the point that it’s barely big enough for myself. I didn’t know that could happen, and it’s not pleasant. I guess I never realized that this might affect more than just my body.

As traffic crawls past the bus stop, I’m overwhelmed by a surge of desperate gratitude—for my car, for the credit card I’ll use at McDonald’s, for my low-interest mortgage, for decades of regular dental care, for my college degree in two impractical subjects, for my husband’s ability to pay our bills while I try to shoot the moon, for my naive ideas about “normal,” for my ability to shrug and think, Fuck Darryl. I came to SDF8 to try to understand what it feels like to work in a fulfillment center. But the thing I really and truly understand now is that, regardless of how broke I may be, I’m the upper class. I always will be. I won’t ever really understand what it feels like to work here, because I know that I get to leave.

If this were a completely accurate representation of my month at SDF8, the next forty pages would consist entirely of complaints about constant pain, waking up at 4:00 a.m., walking between thirteen and sixteen miles per day, being too tired to talk to anyone, eating a lot of McDonald’s, rarely seeing the sun, and passing out the minute I get home from work.

And more than anecdotal evidence suggests that, though American jobs have gotten safer, people’s experience of them is at an all-time low. We’re less likely to get killed or maimed at work than ever before, true. But we’re also at record lows for job satisfaction, job security, free time, and feeling in control of our lives. For just one example, a recent study by the Harvard Business Review and the Energy Project surveyed twenty thousand workers across twenty-five industries and found that American workers have never felt unhappier or less secure at work, with only 37 percent satisfied with their jobs. The authors wrote: It’s a depressing but undeniable reality: the vast majority of employees feel depleted, diminished, disenfranchised, demoralized, and disengaged at work.… For 200 years, since the dawn of the Industrial Age, the model for how to work has been the machine, and more recently, the computer.… Machines are valued for their speed, efficiency, and predictability. They make no demands. When they break, they can be repaired or replaced. Computers run even faster and do more. The assumption in organizations has been that people ought to be able to operate in the same way.

Frederick Winslow Taylor was born into a wealthy Philadelphia family in 1856. Expected to go to Harvard, he instead chose to go into industry by taking a job as a lowly machinist’s apprentice at Midvale Steelworks. As young Fred worked his way up to head engineer, he observed widespread soldiering among his coworkers. “The natural laziness of men is serious,” Taylor would later write in his 1911 magnum opus, The Principles of Scientific Management, “but by far the greatest evil from which both workmen and employers are suffering is the systematic soldiering which is almost universal.” He really meant “evil,” too. Taylor was an early true believer in the idea of efficiency and productivity as savior. Increases in labor productivity at factories employing his “scientific management,” he believed, would raise the wages of the workers, which he saw as the only realistic way of ending poverty and class conflict. Scientific management, or Taylorism, was a new system of eliminating all mystery from factory work, whether it was how fast workers were really capable of going or the “sorcery and black magic” methods of skilled workers. To do this, Taylor used a relatively new technology—the affordable, accurate stopwatch—to measure and analyze previously subjective tasks. It was the birth of literal micromanagement. First, Taylor would pull the best, strongest “first-rate men” and pay them extra to work as fast as possible while he watched, timing their motions to fractions of a seco...

That’s another thing about Fred Taylor—he often claimed to respect and feel kinship with the men he’d labored alongside during his apprenticeships, but he constantly talks shit about them, and seems to regard them as almost another species: One of the very first requirements for a man who is fit to handle pig iron as a regular occupation [is] that he shall be so stupid and so phlegmatic that he more nearly resembles in his mental make-up the ox than any other type.… He is so stupid that the word “percentage” has no meaning to him, and he must consequently be trained by a man more intelligent than himself.

Journalist Upton Sinclair, whose The Jungle was the catalyst for reforms in the meatpacking industry, wrote a horrified letter to American Magazine in response to its publication of the excerpted Story of Schmidt. “I shall not soon forget the picture which he gave us,” wrote Sinclair, of Schmidt and his fellow workers at Bethlehem Steel, “induced to give 362 per cent more service for 61 per cent more pay.” Why, he asked, should they “receive $1.85 for the work, instead of, say, $2.85”? Taylor responded in a letter to the same magazine the next month, explaining that in his experience, if you pay workers too much, they “work irregularly and tend to become more or less shiftless, extravagant and dissipated. Our experiments showed… that for their own best interest it does not do for most men to get rich too fast.” The less obvious but equally important effect of Taylorism was breaking the power that skilled workers held over the pace of work and the operations of factories. He began disassembling skilled jobs into multiple smaller unskilled jobs, often designing specialized tools to replace a skilled worker’s “sorcery and black magic” knowledge base. There was no mystery in the “one best way,” and that removed the job security of skilled workers, further shifting the balance of power toward factory owners.

What’s less well known about The Wealth of Nations is that Smith also describes the ill effects of monotonous, highly divided labor on the day-to-day experience of workers: The man whose whole life is spent in performing a few simple operations… has no occasion to exert his understanding or to exercise his invention in finding out expedients for removing difficulties which never occur. He naturally loses, therefore, the habit of such exertion, and generally becomes as stupid and ignorant as it is possible for a human creature to become. The torpor of his mind renders him not only incapable of relishing or bearing a part in any rational conversation, but of conceiving any generous, noble, or tender sentiment, and consequently of forming any just judgement concerning many even of the ordinary duties of private life.… But in every improved and civilized society this is the state into which the labouring poor, that is, the great body of the people, must necessarily fall, unless government takes some pains to prevent it.

Smith’s theory of markets works on the assumption that the profits of higher labor productivity would obviously be passed along to workers as higher wages—giving them an incentive to put up with the monotony of subdivided work. And in the late 1700s, that might have worked. Businesses were much smaller back then—Smith’s pin “factory” had just ten workers, who’d be doubly motivated by working alongside a boss they knew personally and sharing in the profits. That system has a ceiling, though. Once a company grows to employ more than a couple hundred workers in more and more complicated jobs, it becomes increasingly difficult to know all of them personally, much less keep track of and control them all. The more efficiently you subdivide job duties, the unhappier your workers become. And a workforce of miserable, bored, hard-to-supervise workers quickly leads to soldiering, or situations like the shipworkers playing cards all day while one guy makes a lot of noise with a hammer. At the turn of the century, growth and division of labor were approaching that point of diminishing returns.* But Taylorism had no ceiling. Its combination of objective-seeming data analysis, specific productivity goals, monitoring, and deskilling was the system that growing factories had been desperate for. By conceiving of workers as numbers in an equation rather than individual humans, Taylor made it possible for companies to expand enormously, employing thousands and even millions of workers withou...

Taylor’s results could be incredible, but he left a trail of discontent behind him. Men complained of overwork, exhaustion, and the mind-numbing monotony of this new kind of work. “No tyrant or slave driver in the ecstasy of his most delirious dream ever sought to place upon abject slaves a condition more repugnant,” said one very dramatic labor leader in the midst of a 1911 strike at Boston’s federal Watertown Arsenal. It had been catalyzed by Taylor’s pronouncement that a molding process usually taking fifty-three minutes would now take less than half that long. But even more than the stopwatches, Watertown workers deeply resented Taylor’s inherent premise that they were lazy morons who would just screw around or mess things up if given any freedom at all. They wrote in a plea to their boss: “It is humiliating to us, who have always tried to give the Government the best that was in us. This method is un-American.” Taylorism nominally fell out of favor in the 1930s, and in modern parlance the word has a negative connotation along the lines of “eugenics” or “lobotomy.” But Taylorism’s central premise that workers are lazy, stupid, and never to be trusted became the undercurrent of American management, from Ford’s assembly lines to Taco Bell’s decree that every car at the drive-through be greeted within four seconds of pulling up.

The technology available today allows for the kinds of individual, minute levels of supervision and control Taylor could only dream of. The digital ghosts of his stopwatches and “one best way” have infiltrated previously hard-to-quantify sectors of the US workforce, digitally deskilling and speeding up work. Techno-Taylorism has gotten the most attention as it starts making inroads into high-wage jobs such as law, medicine, and journalism, but it’s already a way of life in the low-wage sector—fast food, call centers, nursing, elder care, etc. It’s very difficult for modern unskilled workers to avoid working in a techno-Taylorized environment defined by mistrust and contempt. This is because technology also multiplies Taylorism’s second prong—deskilling. “If you don’t know what to do, just check your scanner—it’ll tell you” is Michelle’s golden rule, and it’s true. Nearly every drop of “brain work” has been wrung out of this job, like so many others. Cashiers used to have to be able to do math in their heads to give people change; today, many registers automatically dispense the appropriate coins. Call-center workers rely on elaborate scripts instead of understanding the systems they work within. London’s taxi drivers, who once had to study for years to pass the Knowledge, are being displaced by Uber drivers with GPS maps. As more and more skill is stripped out of a job, the cost of turnover falls; eventually, training an ever-churning influx of new unskilled workers become...

It’s not the injuries and physical pain, Walker says. It’s more the stress, and the despair. “When you’re working in a life role with high demand and very minimal resources, you get very high levels of toxic kinds of stress. The job that you’re in is classic—you have no control over the demand that’s placed on you for productivity for any given hour that you work there. That is a contributing factor to the innate biological desire for relief. And that desire is part of what motivates the substance abuse,” he says. I’m fascinated. “You know, since I started, I’ve—you know everyone’s inner four-year-old that just wants McDonald’s?” “Yup!” “Since I’ve started this job, I have a lot less control over my four-year-old. Like—‘Whatever, screw it, I’m going to McDonald’s.’ Is that related?” “That’s right!” he says, excited. “That is exactly the same thing! Some people will turn to cigarette smoking, others it’s overeating, others it’s over-drinking, for others it’s drug use.” Then he says something I haven’t been able to get out of my head to this day: “The human mind is designed to not feel bad. It does not like to feel bad. And it will tend to do things that it can to correct for feeling bad. And, unfortunately, things like drugs, alcohol, tobacco, and food—particularly carbs—make us feel better. It’s a native tendency to self-medicate into a better mental state, to get rid of this overwhelming negativity.

Blair says it’s nice to talk to me even if I am a journalist, because she hasn’t bothered to make many friends during peak. “I’d heard they had a crazy turnover rate because of the workload and the long hours, so I was, like, ‘Don’t get attached.’ I got buddy-buddy with one girl in my training session because I could tell she was serious, like me—but nobody else. And we were looking around at standup yesterday, like, ‘Where is everybody?’ She was, like, ‘Do you see anyone else from our training group?’ And I was, like, ‘No one.’ Not one person from our entire group of—it was over five hundred people.” I raise an internal eyebrow at the idea that Blair could remember the faces of more than five hundred people in her training group. But that level of turnover does sound plausible. Darryl’s* the only other person from training I see around anymore, and I barely recognize any faces at standup after two weeks. The Amazon newsletters posted weekly on the bathroom stall doors have sidebars of fun statistics, including how many new hires there were the previous week. Even during these final few weeks of peak, it’s been around twenty-five new blue badges and between four hundred and five hundred new white badges hired every single week. A 99 percent turnover rate for pickers since Thanksgiving isn’t that implausible.

Mostly, though, she takes pride in being good at her job. Her numbers, she says, are usually double or triple what other pickers can manage. That doesn’t happen for her at every job—she actually technically qualifies for disability. “Between my Asperger’s, which is technically classified as autism spectrum disorder, and hearing loss and ADHD and PTSD—that lovely storm of me—that’s what qualified me,” Blair says. “But I just couldn’t accept a handout. I love knowing that the government had my back, but I was, like, ‘My mind might be a little special, but my body’s fine. I can physically work.’ “And I love that I’ve found a job that actually utilizes the way that I am. My eidetic* memory—I’ve never really been able to use that at a job before. Like, at the grocery store, I remembered all the codes, even for other departments that weren’t mine. They’d be coming to me, like, ‘What’s the code for artichokes?’” Blair laughs. A photographic memory is very useful at SDF8, she says. When an item code she’s encountered before comes up on the scanner, “there are images attached, not just numbers. I’m remembering the bar code, and its picture, and its feel—that’s one of the things that makes me so fast.”

In 2012, Amazon bought Kiva Systems, a robotics company specializing in automating the process of picking and packing in warehouses, for $775 million. SDF8, which opened in 2012, is “seventh generation.” The eighth-generation fulfillment centers that opened after 2014 integrate robot and human workers to make picking a stationary job. The robots are orange and kind of cute, like bigger, stronger Roombas balancing drawers on their heads. “The video I saw, they’re, like—shelves that move,” says Blair. “They move and dance about the warehouse and communicate with one another; they bring the shelf that has the right item to a human picker. They even face it the right way, so all you have to do is—BEEP!” She perfectly reproduces the noise of the scan gun we hear thousands of times a day.

John Henry is an American folk hero, possibly based on a real African American railroad worker who died helping to dig the Chesapeake and Ohio Railway’s Big Bend Tunnel in West Virginia in the 1870s. At the time, tunnels had to be manually cut through rock. The biggest and strongest railroad workers, called steel-driving men, would use heavy hammers to pound long steel rods as deep as twelve feet into the rock of a mountain. The holes were then filled with explosives and detonated. In the legend, John Henry was the biggest and strongest of the steel-driving men. He supposedly wielded two twenty-pound hammers, one in each hand. One day, the railroad brought in a steam drill to replace the steel-driving men. John Henry, out of pride or an instinct for job preservation, challenged his boss to let him race the machine, one on one. The boss agreed, and all the railroad workers gathered to watch the battle between man and machine. John Henry and the steam drill were neck and neck for hours, but as they neared the finish line, the drill finally broke. John Henry kept going and won the contest, but he’d pushed himself too hard—as he struck the winning blow with his hammers, his heart gave out and he fell over, dead. The folktales never really get into what happens after that, but history would suggest that the boss repaired the steam drill and fired the remaining steel drivers as if the contest never took place.

It suddenly starts to rain, and most of the guys run for shelter in the RV park’s men’s room. I take cover under the kitchen tarp with Eli. “I’ve been traveling since I was nineteen,” he says. “Not like how these guys do, hopping trains and shit, but I’ve been to nineteen countries. I see jobs as a means to my end. I make probably ten grand a year, and I feel like I’m living like a rich man. How many millionaires go on five-month vacations? I’m giddy as hell about the next five months. I have no idea what it’s going to bring—I have an entry flight and an exit flight and three countries to explore.” It sounds amazing, and I tell him so. “I retire every year! Why wait until you’re sixty-five? I don’t know if I’m gonna be in good health—my brother’s fifty, and he’s falling apart! I’m gonna live every year the best I can. There’s so many people stuck in their ruts—not happy, going to work to pay bills for houses and cars they can’t afford. Taking money way too seriously makes them hate their jobs, and then they hate their life.” What about kids? I ask. Eli shrugs. “I don’t have any kids. And I don’t have a house. I have no bills—well, I have my cell phone and the insurance and upkeep on my van, but the job in Alaska takes care of that for the year, and Amazon is for my flight and some spending đng. I feel like I’m living a dream, and I never want to wake up.”