Most companies don’t boast about their penchant for firing people. Netflix is a different story. One of the core tenets of Netflix’s workplace culture is that it functions not like a family, but like a professional sports team. That means anyone—no matter how hard-working they are, or how good their track record might be—is vulnerable to getting cut.
That principle was on full display this month with the surprise ouster of Cindy Holland, Netflix’s longtime vice president of original content, who was responsible for developing a powerhouse slate of domestic programming, from House of Cards and Orange Is the New Black to Stranger Things.
Nonetheless, Netflix co-CEO Ted Sarandos initiated a restructuring that left Holland without a role at the company, and named Bela Bajaria, who had previously overseen non-English TV programming at Netflix, the new head of global TV. As the company looks for more growth by expanding further into international markets, Holland apparently didn’t pass the Keeper Test.
What’s the Keeper Test, you ask? It’s the scenario that all managers at Netflix are encouraged to ponder, as Reed Hastings, the streaming giant’s co-founder and co-CEO, explains in his new book No Rules Rules:
If a person on your team were to quit tomorrow, would you try to change their mind? Or would you accept their resignation, perhaps with a little relief? If the latter, you should give them a severance package now and look for a star, someone you would fight to keep.
There’s no denying that Netflix is a hugely successful company, with 193 million subscribers and a valuation of $223 billion. But is that because of the Keeper Test, or in spite of it?
In No Rules Rules, written as a dialogue with organizational behavior professor Erin Meyer at INSEAD, Hastings says the company’s Darwinian approach to talent is necessary to maintain a competitive edge.
Hastings has suggested that business leaders in “a creative industry where there’s a lot of change” may well be able to extrapolate lessons from Netflix for their own organizations, and his book was recently shortlisted for the Financial Times and McKinsey Co. Business Book of the Year Award for 2020. Between Hastings’ ability to amplify his message and the influence of Netflix’s workplace culture overall—its original culture slideshow deck has been viewed close to 20 million times—understanding how the Keeper Test impacts people matters for employees at all kinds of companies.
It’s worth noting that turnover at Netflix isn’t actually that high. In 2018, it reportedly had an 8% firing rate—above the 6% average for American companies, but not exorbitantly so. Netflix also had far less voluntary churn (i.e. people choosing to leave the company) than is usual for US corporations. As the Wall Street Journal has reported, Netflix’s 4% churn rate in 2018 compared to a 13% average for American companies, as measured in a 2017 report by the Society of Human Resource Management.
But the Keeper Test looms large in company lore and in the minds of employees. So even if the risk of losing a job at Netflix isn’t that much higher than it would be elsewhere, it can certainly feel that way for people who work there. “I think some people felt it was a culture of fear,” a former vice president of talent at Netflix told the Wall Street Journal in 2018.
From a practical point of view, there are at least three big potential issues with the Keeper Test.
The first is whether employees might be so worried about the prospect of losing their jobs that their performance suffers. Research suggests that job insecurity may lower employees’ satisfaction and creativity, as well as heighten their risk of depression; it’s generally hard to focus on doing your best when you’re waiting for someone to tap you on the shoulder and tell you that you’re out.
In No Rules Rules, Meyer speaks with one director at Netflix who says he didn’t unpack his boxes during the first nine months in his job. On job-review site Glassdoor, Netflix employees past and present frequently mention living with anxiety that they’ll be let go. “While working at Netflix it feels like you’re trying to survive rather than thrive. Your job is always on the line, you can be let go at any point and people work in fear,” reports one reviewer.
But that same reviewer nevertheless says they’d recommend working at Netflix. In fact, overall, Netflix employees on Glassdoor give the company an average four out of five stars, citing its great pay, benefits, autonomy, smart people, and perks. For some people, lower job security is worth the tradeoff—and that’s just the sort of employees that Netflix wants to bring on board.
“We have to hire the psychological type that can put [fear] aside and who aspires to work with great colleagues and that that’s their real love, is the quality of their colleagues or the consistency of that, versus the job security,” Hastings said in a recent interview with Variety. He emphasizes that the Keeper Test isn’t meant to freak people out, but to ensure that the best possible person is in every role. “I find it motivating that I have to play for my position every quarter,” he writes in No Rules Rules.
Of course, the possibility of getting fired isn’t so terrible when you’re a powerful CEO with a multibillion-dollar safety net. But to the company’s credit, it’s upfront about the Keeper Test and its firing tactics. People who go to work there know what they’re signing up for, and that they’ll get a generous severance package (four months minimum) if they do get shown the door. They also know that having the company on their resume, even if only for a brief tenure, makes them more competitive in the larger job market.
So while the Keeper Test does provoke anxiety, the majority of Netflix workers seem to find that it doesn’t dominate their experience at the company—perhaps because Netflix is adept at hiring people who are willing to put up with it.
The second potential problem with the Keeper Test is that it could incentivize some very counterproductive behaviors—from excessive caution to constant in-fighting and undermining.
Meyer writes in No Rules Rules that her initial reaction to hearing about the Keeper Test was that it would stifle psychological safety—the principle coined by Harvard Business School professor Amy Edmondson that “if you want to encourage innovation, you should develop an environment where people feel safe to dream, speak up, and take risks.” How safe can people feel if they’re always worried about getting on their boss’s bad side?
But Netflix seems to have gotten around that problem by simultaneously encouraging blunt feedback and risk-taking. “At Netflix, it is tantamount to being disloyal to the company if you fail to speak up when you disagree with a colleague or have feedback that could be helpful,” Meyer writes. “After all, you could help the business-—but you are choosing not to.” Managers put requests for feedback on their agendas for one-on-one meetings with employees and are the recipients of “360” performance reviews that include constructive criticism from their own reports.
As for risk-taking, Hastings writes that the mantra at Netflix is that “employees don’t need the boss’s approval to move forward (but they should let you know what’s going on).” By hiring smart people and then handing them lots of decision-making power, the company aims to cultivate an atmosphere where employees feel they are more likely to lose their jobs for turning in a safe, adequate performance than for taking a big risk that doesn’t pan out.
What about backstabbing? “We are very careful not to have any firing quotas or ranking system,” Hastings says, noting that systems that encourage managers to compare employees against one another can discourage collaboration and lead to cutthroat, toxic environments. Meyer also says that she saw little evidence of Hunger Games competition afoot in her interviews with employees. But one former Netflix employee told the Wall Street Journal that when she was let go, her boss suggested that one thing she could have done differently was to fire someone on her own team sooner. (Netflix denies that the employee received this feedback.)
Hastings suggests that the Keeper Test works because “everyone at Netflix is happier and more successful when there is a star in every position.” But is it desirable, or even possible, to have a team made up exclusively of stars?
For one thing, it’s unclear what constitutes a star employee in Netflix terms. That means the Keeper Test may be vulnerable to misjudgment or bias on the part of managers. Research shows, for example, that we tend to have better evaluations of people who remind us of ourselves, and that favoritism often determines career advancement or lack thereof.
Netflix says managers are expected to inform employees who are falling short, so that they’re not surprised if they get the ax. But some former employees say they didn’t receive feedback or opportunities to improve.
Research suggests that star performance may be determined not just by a person’s innate talent, but by environmental factors, such as how much access a person has to organizational resources or the strength of their social network, as Ernest O’Boyle and co-author Sydney Kroska write in the Oxford Handbook of Talent Management. There’s also evidence, they write, that “stars’ output is attributable as much to their network and support structure as it is to any internal attributes.”
But the Keeper Test, as explained in No Rules Rules, suggests that Netflix ascribes to the “great man theory,” wherein a person’s performance is both entirely under their control and a reflection of their efforts alone.
No one would argue that it’s wise to hang onto underperformers. But a company that writes off people in the way Netflix does—Hastings endorses “firing a good employee when you think you can get a great one”—may not be devoting sufficient time to the question of how they might invest in good, hardworking employees to help make them great.
Research shows that subpar performance can often be traced back to a manager who has created an unhealthy dynamic via low expectations or lack of confidence, as Jean-François Manzoni and Jean-Louis Barsoux explain in their book The Set-Up-to-Fail Syndrome. “Having made up his mind about a subordinate’s limited ability and poor motivation, a manager is likely to notice supporting evidence while selectively dismissing contrary evidence,” the pair write in Harvard Business Review.
In No Rules Rules, Hastings downplays the potential pain caused by firings. For example, discussing his dismissal of former Netflix chief talent officer Patty McCord, who helped create the Keeper Test herself, he writes:
“I started feeling that it would be best for us to have someone new in the role. I shared those thoughts with Patty, and we talked about what was leading me there. As it turned out she wanted to work less, so she left Netflix and it was very amicable.”
McCord would agree that she and Hastings are still close, but her recounting of events is markedly different. “What I was surprised by was the profound sadness that I wouldn’t be part of it anymore,” she told Alex Blumberg on the podcast Without Fail. In McCord’s telling, the ouster wasn’t a completely mutual decision, nor does she describe it as a convenient way for her to work less. She was hurt, as most anyone would be.
Indeed, Netflix often seems to try to skirt the question of the mental toll of getting fired, a position that McCord herself takes in a 2013 interview with NPR’s Steve Henn:
HENN: How many people do you think you have fired?
MCCORD: Oh, I would really like to remove that word from our vocabulary. It’s like, we don’t shoot people.
HENN: OK. What word do you like? Severed is no less pleasant than fired.
MCCORD: No. You just move on.
HENN: How many people have you moved on?
MCCORD: Oh, hundreds.
McCord’s preference for the phrase “move on” rather than “fire” may well be intended to reflect the fact that Netflix employees aren’t necessarily let go because of poor performance. At the same time, it suggests a certain refusal to consider the emotional toll of the Keeper Test too deeply.
Ultimately, the Keeper Test is the result of a mindset that employees are disposable. In this sense, Netflix isn’t so different from a lot of other American companies. What’s new about the Keeper Test isn’t so much the end result (people getting fired, or worrying about the prospect of getting fired) but rather the company’s position that the test is a good thing, and even a selling point.
In The End of Loyalty: The Rise and Fall of Good Jobs in America, author Rick Wartzman writes about how the post-World War II social contract that once existed between workers and their employers was gradually undone by the idea, brought to the fore by the economist Milton Friedman in the 1970s, that a business’s primary obligation was to its shareholders. Wartzman describes the shift as a move away from “this kind of ‘we’ mentality—we’re all in this together coming out of the Great Depression and the Great War—and moving more to an age of I, the individual,” as he told NPR. “The marketplace is sacrosanct, and it’s all about, you know, lifting ourselves up.”
There are echoes of this in McCord’s framing of the culture she helped create: “Companies don’t exist to make you happy. You know that, right? The business doesn’t exist to serve you. The business exists to serve your customers,” she told Fast Company in 2016.
But while companies don’t exist to make employees happy, the growing white-collar labor movement and the rise of conscious capitalism in the business world suggest that the relationship between companies and their workers may once again be on the brink of change. Even powerful CEOs like JPMorgan Chase’s Jamie Dimon and Walmart’s Doug McMillon are at least paying lip service to the idea that the purpose of companies isn’t just to serve customers or shareholders, but to take the interests of employees and communities into account, too.
The Keeper Test is the logical product of a culture where job security is largely a relic of the past. Netflix’s high pay and severance packages doubtless mitigate some of the damage. But they don’t supply an answer to the moral argument that, given what we know about the impact that unemployment has on people’s mental health—and, in the US, on their access to health insurance and ability to find a new job—the bar for firing people should be high, not low.