Amazon has built its $1.3 trillion empire largely by tracking and evaluating almost every aspect of a customer's life. From a new TV to a toilet-paper refill, Amazon knows what a customer wants and when they want it, and it's always ready to serve it to them.

This obsession with metrics and data, however, does not appear to extend to certain parts of Amazon's workplace. Over the past few months, the company has aggressively pushed employees back to the office. In February, Amazon announced that employees would be required to come into the office three days a week and since then, the e-commerce giant has escalated its battle with remote employees: sending emails to employees about their attendance, creating internal dashboards to display how many days a week each employee was coming into the office, and telling managers in October that they could begin firing employees who weren't meeting the return-to-office requirements.

When perturbed employees have pressed executives for the reason behind the mandate, supposedly data-obsessed higher-ups have seemed to have no data to justify it. Asked in August about this, Mike Hopkins, a senior vice president of Prime Video and Amazon Studios, offered a vague response, saying that he had "no data either way" on whether mandating in-office work made people more productive but that executives believe Amazon's workers do their best work when they're together. 

It's reasonable to wonder why Amazon, a company that has data on hundreds of millions of people and their decisions, is struggling to come up with hard numbers to back up its dictatorial push back to the office. Perhaps the reason is that the data supporting Amazon and other companies' RTO policies is threadbare, relying mostly on a few studies that use sample sets of questionable usefulness to back up their claims that remote work is less productive. 

But the weakness of the evidence won't stop bosses from making these RTO mandates. They indicate a failure of imagination on the part of management and a refusal to do the work necessary to create a positive company culture.

Evidence is as evidence does

As the return-to-office battle has heated up in the past six months, there has been a marked increase in declarations that remote work is less productive. But diving deeper into this evidence reveals flawed logic — and a media industry obsessed with proving bosses right.

The study that has most often been used to argue for the necessity of in-office work is a July working paper from researchers at the National Bureau of Economic Research, which randomly assigned data-entry workers at a company in India to work either from home or in the office for eight weeks. The researchers determined that remote workers were 18% less productive than their in-person counterparts. Journalists have consistently cited this study without, it seems, taking a moment to consider its findings. First and foremost, it's farcical to use the work of entry-level data workers in India — who were recruited specifically for the study — as a proxy for all employees in all industries around the world. Secondly, the measurement of their productivity was "net speed," or the number of correct entries they made in a minute. The "drop in productivity" is really about how fast people could put numbers into a sheet — but that's not what most people do at work. 

Other outlets cited a study that examined the productivity of 10,000 workers at an Asian IT-services company. The central claim seems to be tailor-made for RTO advocates: The researchers estimated a productivity shortfall of 8% to 19% when workers transitioned from working at the office to working from home. The study's authors found only a "slight decline in output" but acknowledged workers were stretching out their working hours at home; those things together showed up as a drop in productivity. Here too, there are issues. "Output" in this case refers to "performance against the semi-annual goal on a key performance metric," like lines of useful code written by a developer. The authors described these metrics as "objective" and tracked, but the actual composition of these metrics is still pretty vague and based on managers' interpretations of their value. Additionally, hours worked were logged by an employee-surveillance software — but these sorts of tools come with plenty of problems that should make anyone skeptical of their efficacy.

These studies — and the RTO push — often betray an utter ignorance of the workplace and work itself, both its structure and its outputs. 

You'll notice a lot of these studies focus on call centers (the Stanford researcher Nicholas Bloom cited two in his roundup of productivity research), and that's likely because these are extremely controlled and heavily micromanaged environments — ones rife with labor abuse. Crude measures of productivity might indeed slip when workers are able to get away from horrible managers or torrents of abuse, but "productivity" in these studies is always a rigid metric, like "calls answered," rather than something more meaningful, like whether a problem was fixed or whether the customer was happy. These studies are relatively useless when it comes to evaluating most companies' return-to-office strategies, but that's just fine for the managerial elite.

As somebody who's been writing about this subject for years (and who's worked remotely since 2012), I've yet to read a single piece of research that convincingly backs up the assertion that we need to be in the office. And yet major media outlets have continued to feed bosses' narrative that we "do better work together." The existing studies continually fail to evaluate real work. Instead they prioritize speed, betraying the same corporate ignorance forcing people back to the office. These studies — and the RTO push — often betray an utter ignorance of the workplace and work itself, both its structure and its outputs. 

Just the vibes

Despite the limited evidence against it, corporations are increasingly trying to kill remote work. Corporate statements about these decisions never seem to justify the shift beyond platitudes about "togetherness" and vague references to "culture." Look deeper, though, and they reveal the rotten core of the mandates.

Roblox, a company that derives its revenue from digital worlds, reversed its flexible work policy in October, telling employees that if they didn't work at least three days a week in the office they'd be laid off. The reason for the about-face? Based on its CEO David Baszucki's blog post announcing the move, it seems mostly about vibes.

"I personally hoped that for our culture and our type of work, it might be possible to imagine a heavily hybrid remote culture," he wrote. "But there was a pivotal moment for me when we had our first post-quarantine, in-person group gathering. Within 45 minutes I came away from three separate conversations with spontaneous to do's and ideas to put in motion, something that hadn't happened during the past few years of video meetings."

Nike's four-days-a-week policy points to "the power and energy that comes from working together in person." Geico said its return-to-office program was meant to "foster a sense of community and connection" — but offered little data to support the decision. The only numbers in the announcement appeared to be about layoffs: Geico said it was letting 2,000 people go to sustain "long-term profitability and growth," suggesting that while layoffs can be evaluated with data, in-person work can only be backed up with mood rings.

Managers and executives make calls based on perception rather than hands-on experience or data.

These announcements are almost always issued by executives who probably won't be subjected to the same kinds of check-ins as the rank-and-file workers the mandates apply to. Nobody's asking Amazon's Andy Jassy or Geico's Todd Combs how many days they swiped into the office, and there's no chance Oracle, which instituted a return-to-office policy in May, would punish Safra Catz or Larry Ellison for spending too little time at their desk. This irony combined with vague justifications exposes the reality of the RTO push: Managers and executives make calls based on perception rather than hands-on experience or data. The modern CEO has become a figurehead reaping the rewards of a work process they don't meaningfully participate in, so they make their choices based on macroeconomic conditions, their own biases, or, evidently, a single 45-minute meeting that left them feeling good. 

Nowhere is this more obvious than at Meta, where workers are required to return three days a week. The only problem: Employees can't find the space or the privacy to actually do their work at the office. The issue is almost too on the nose. The honchos calling the shots at Meta, a company that has caused global discord through its handling of people's data, appear to be unaware of how people at the company do their jobs. More frustratingly, Meta's corporate guidelines say that anyone who has worked at the company for 18 months or more can apply to become a permanent remote worker — a nice idea, except for the fact that hundreds of people have applied and have yet to hear back. 

Because I said so

Creating and sustaining a positive, productive corporate culture takes work. It requires managers and executives who truly understand the product their employees are putting out — and what is required to create that product. That's what makes the move to kill off remote work so frustrating. It's not clear that the return-to-office move is about making workers more productive or building a better culture. Rather, it's becoming obvious that these mandates are mostly an attempt to reestablish a surveillance society that allows managers to skip over the tough task of building a company where people actually want to work.

The RTO push is eyewash for investors to prove that drops in revenue and profitability aren't a result of poor managerial decisions but the result of lazy workers sitting at home in their pajamas. In some ways, it's a genius move for executives — a way to establish control over workers during an unprecedented societal awareness of labor rights (thanks to the striking workers of the Writers Guild of America, SAG-AFTRA, and the United Auto Workers) while also shifting the blame and consequences of poor stock performance onto those least responsible.

Real management takes responsibility and makes thoughtful decisions based on what makes a company stronger.

I would perhaps have more sympathy if companies made even the lightest attempt to demonstrate the efficacy of office work through relevant data or definable productivity metrics, rather than vague references to hours worked or office attendance. But seeing this type of data is unlikely as corporations have mostly turned modern managers into hall monitors. And if, as some have suggested, the return-to-office push is an attempt at a "soft layoff" — instituting unreasonable policies to make people quit (or accept severance) — it's corporate cowardice. It's restructuring a company based on who's most willing to tolerate wrongheaded inconveniences, rejecting great workers who don't live close to an office, and galvanizing sycophants who are willing to uncritically cheer on every executive mandate. 

While executives may see a return-to-office push as a good thing, I believe these mandates will only weaken their organizations, driving a wedge between management and workers. Forcing in-person attendance without clear goals and reasoning is going to create an outright hostility toward higher-ups. This pointless and petty crusade does nothing to make organizations better, leaner, or more productive — all it does is temporarily help executives distract from larger organizational issues. 

Real management takes responsibility and makes thoughtful decisions based on what makes a company stronger. The return-to-office move is the exact opposite: an unproductive push for control that erodes the already tenuous loyalty workers have to their employers while failing to address core problems of managerial competency that will only get worse as decent employees flee these dimwitted demands.

Ed Zitron is the CEO of EZPR, a national tech and business public-relations agency. He is also the author of the tech and culture newsletter Where's Your Ed At and the host of the "15 Minutes in Hell" podcast.