Against performance fetishization

Let’s start with a story. Consider a tech worker at a tech company, one of the many employees caught up in a modern “high-performance” culture. The company hires a new director to “transform” the department. You know, the kind of transformation that comes with big promises of “innovation” and “disruption”, in favor of whichever management theory is popular at the time.

As it soon turns out, this shift results in the arbitrary halting of ongoing work, concurrent initiatives resulting in duplicated efforts, and constant messages in multiple communication channels generating a maze of contradicting expectations. The frequent ambiguity and shifting of priorities prove harmful to both job quality and team well-being. Perhaps our tech worker should do a better job of guarding their own priorities and being more selective about what to focus on, but the reality is that the entire team is caught up in this maelstrom of productivity theater, where everyone is constantly pressured to “show performance” and “prove their productivity”, in the expense of actual work.

This situation goes on for months, and throughout the duration, the entire team is subjected to scrutiny, bound in an endless loop of trying to establish their worth by arbitrary metrics and deliverables, while the actual value of their work appears increasingly meaningless. Each week comes additional demands to measure their efforts and demonstrate that they are not simply just occupying space. In that light, when the tech worker of our story hits their breaking point, which leads to burnout, they discuss this with their manager. They expect for understanding or at least for some sort of awareness of the untenable condition. Instead, their manager responds with a complete lack of empathy or acknowledgment of the situation, followed by the useful insight that “things will only get worse next year”. 

This isn’t just a story about poor managerial practices, everyone knows there’s an abundance of those… It tells the story of what occurs when companies build layers of management whose primary function appears to be justifying their own existence by elevating themselves as judges over others. 

In modern “high-performance” cultures, burnout is not viewed as a systemic signal for concern, but rather as an individual shortcoming. A burned-out worker is viewed by management as either weak, willfully disobedient, or simply lazy. Burnout is viewed as a performance problem and an indicator that someone is “checking out” of the organization. Burnout is rather a stain on the employee’s record, a mark on their “performance evaluation”. This stigma response acts as a potent deterrent because it uses a burned-out employee as an example and a warning to others, telling them to be quiet, keep their heads down, and hide their fatigue, or else…

On an individual level, this reality results in exhausted tech workers who are pressured to complete progressively pointless tasks to meet arbitrary goals. At the product level, it leads to what we now refer to as the “enshittification” of technology, as once-helpful tools turn against their users more and more to meet growth demands. Furthermore, it pushes us in the direction of ecological catastrophe on a planetary level since the constant need for greater engagement, data, and processing capacity results in rising resource usage and thus degradation of the environment. 

The rot goes beyond poor product choices or poor management as we are all expected to continue “business as usual” in a time of polycrisis. While dealing with the trauma of a lingering global pandemic, confronting the reality of climate breakdown, and navigating economic and geopolitical instability, workers are expected to maintain peak productivity. Having to take part in performance theater—attending happy status meetings, attaining arbitrary KPIs, and celebrating “wins” while the world seems to be falling apart is incredibly disorienting and a unique sort of corporate gaslighting. Burnout is one manifestation of this collective cognitive dissonance.

In the end, I know this story intimately well. And that is because I was the tech worker of the story. The sad part is that my experience of burnout is not unusual; rather, it is a microcosm of a broader trend in which conducting business as usual has become essentially unsustainable in multiple aspects of our lives. And it seems to get worse.

The rise of the performance industrial complex

Something peculiar happened in the 20th century. As actual productive activity got more mechanized and efficient, we saw the rapid expansion of administrative bloat. This wasn’t just a few additional middle managers or HR staff but rather, whole sectors arose devoted to tracking, assessing, and improving worker performance.

Take the contemporary corporate performance review as a prime example. What began as a somewhat simple method of assessing employee contribution has evolved into a complex system of mandated rankings, competency frameworks, and measurements. These days, businesses spend budgets on consultants, software platforms, and training initiatives devoted to the elusive “performance management”.

The problem of course is that we are not attempting to figure out how to actually make workers productive in a meaningful sense. Like say, to improve their work-life balance, or just figure out how people could work for fewer hours. Rather, we are simply developing more complex methods to quantify and record their work. The McKinsey method of gauging developer productivity by lines of code is a prime example of this. The result of this performance fetishization is a form of what David Graeber called “bullshit jobs“. Whole professions of project managers, scrum masters, and “transformation specialists” are devoted to overseeing, evaluating, and reporting on the work of others instead of engaging in any creative activity themselves.

This system has the ability to produce its own self-justifying mechanisms. Think about it for a second… In a vicious cycle, poor performance measurements result in the hiring of performance management specialists, who in turn develop increasingly complex metrics that need to be tracked by even more specialists. As a result, companies are hiring more and more people whose primary responsibility is to oversee the work of others.

The actual human cost

Constant performance monitoring has a serious negative psychological impact. Employees report feeling worried, burned out, and dehumanized. Most of all, they believe that companies do not trust them. As a result, their trust in those companies diminishes. And they’re not wrong. Constant monitoring, and being continuously under pressure to surpass arbitrary standards goes against human psychology and well-being. This shows up in a number of ways. It becomes more stressful to worry about meeting performance goals than to actually work. When people prioritize what can be measured instead of what is valuable, creative and collaborative work suffer. Employees get defensive and conceal their actual work processes, or they simply learn to “game” the system, to satisfy fictitious criteria. 

In the recent past we have witnessed a significant exodus of workers (also known as “The Great Resignation“), especially after COVID. People finally chose to leave the workplace when it turned into a never-ending cycle of performance evaluation and artificial objectives. In the after-COVID era, people are just refusing to engage with a system that views them as machines to be optimized rather than individuals carrying out important tasks. They are neither lazy nor unmotivated. 

However, there is a darker sibling to this voluntary exodus and a counteraction, perhaps, from the companies themselves (especially proliferated by big tech): the deliberate use of layoffs. Businesses have figured out that mass layoffs accomplish two goals and very efficiently. They first instill terror in the remaining employees, who are then under pressure to take on the workload of their laid-off coworkers, you know “doing more with less”. Which in turn becomes a survival tactic rather than a managerial cliche. In the end, the labor market is overrun with desperate (and likely dependable) people who are prepared to put up with harsher monitoring and poorer working conditions in order to find work. Businesses successfully use economic uncertainty as a weapon to compel adherence to more stringent performance standards. This is especially starting to make sense when waves of layoffs are frequently followed by record earnings. These seem calculated actions rather than acts of economic necessity. An announcement of thousands of layoffs by Google or Meta sends shockwaves through the sector, increasing the likelihood of other companies to follow suit.

Perhaps most disturbingly, this performance obsession is aligning perfectly with increasingly reactionary politics among tech oligarchs. We see this in Meta’s systematic dismantling of diversity initiatives and fact-checking systems to curry favor with US authoritarian tendencies, or in tech billionaires’ strategic theatrics of “masculine energy” while seeking political advantage. The automation of and defaulting to bias becomes a feature as executives celebrate “progress” and openly pivot away from inclusion efforts that make the usual march of pure nonsense corporate metrics slower. When corporate leaders abandon even the pretense of ethical constraints, we shouldn’t be surprised to see them embrace ideologies that reduce human beings to mere numbers.If you want to see the human cost of this algorithmic control illustrated visibly you don’t have to look further than the gig economy, where workers find themselves at the mercy of ambiguous systems they can neither understand nor challenge. Delivery drivers cluster outside restaurants, mystified by the seemingly arbitrary decisions that determine their livelihoods – why one worker gets chosen over another, why they suddenly lose access to the platform, why their pay fluctuates wildly for the same task. The algorithms controlling their jobs act as a kind of digital feudal lord – omnipresent yet unreachable, capable of terminating employment without explanation or appeal. When drivers try to understand the logic behind these systems, they encounter only automated responses and circular explanations. Their attempts to optimize their own performance become a desperate game of trial and error, as they experiment with different locations and behaviors to please an inscrutable digital overseer.

The ultimate performance fetish

Here’s where it gets worse… Executives now have their ultimate wet dream thanks to the introduction of AI. AI, in their mind, introduces the prospect of continuously boosting output without being hampered by those annoying human demands. The idea of AI replacing human labor while preserving or boosting output seems very alluring to today’s CEOs. The business message surrounding the use of AI reflects this approach. And so, a pattern will become apparent if you pay close attention to earnings calls or if you read between the lines of press releases. Rarely is it about improving working conditions or enhancing human capacities. Instead, it’s all about “efficiency gains” and “productivity improvements” which are corporate euphemisms for accomplishing more with fewer workers.

This dynamic creates a particular relationship between tech workers and AI systems. Engineers, product managers, and designers find themselves in a precarious position as they develop the very tools that could, in theory, eventually replace them while simultaneously being pushed to demonstrate high performance. Perversely, this is genius. They are compelled to participate in a strange form of digital seppuku.

Growing skepticism about tech’s direction, widespread burnout, and increased union organizing aren’t just reactions to current poor working conditions. They simply represent a dawning recognition among tech professionals that the very systems they’re building are being weaponized against them. The surveillance infrastructure they’ve helped construct becomes the blueprint for their own replacement.

AI is the ultimate expression of what management theorist Peter Drucker warned about decades ago – the tendency to prioritize measurable efficiency over actual effectiveness. AI systems excel at optimizing for whatever metrics we give them, but they often miss the broader context and nuance that makes work meaningful and valuable in human terms. They represent the triumph of quantification over quality, of measurement over meaning. What began as tools to augment human capabilities are rapidly becoming replacements for human judgment altogether. Code completion tools become code generation systems. Writing assistants become autonomous content agents. Each advance is celebrated as “progress”, while the fundamental question “progress toward what?”  remains unasked and unanswered. This marks a qualitative shift in how we conceive of work itself – from a fundamentally human activity that might be assisted by machines, to a mechanical process that occasionally requires human oversight. The performance fetish thus reaches perhaps its final destination: the handing over of human work to algorithmic imperatives.

Research is beginning to reveal how these systems don’t merely reflect our biases but they actively amplify them through feedback loops of human-machine interaction. Studies show people are roughly three times more likely to change their decisions when disagreeing with AI compared to disagreeing with other humans, yet consistently underestimate this influence. The implications are chilling: as AI systems learn from human biases embedded in their training data, they magnify these prejudices and feed them back to users, who then internalize and amplify them further. This creates a vicious cycle where both human and machine judgments become progressively more skewed over time. In workplaces increasingly mediated by AI systems, from hiring algorithms to performance evaluation tools, this bias amplification mechanisms serve as a powerful force multiplier for existing inequities – all while maintaining the façade of objective, data-driven decision making. The performance metrics become vectors for encoding and intensifying societal prejudices under the guise of technological neutrality.

Business as usual = capitalist realism

This whole obsession with performance fetishization is the ideal way to illustrate an angle of Mark Fisher’s “capitalist realism”. By capitalism realism, Mark has described the pervasive belief that capitalism is not only the dominant system but the only system we can ever have imagined. The majority of employees now find it easier to envision their personal well-being deteriorating, or them completely abandoning the tech sector (with whatever implications this may have on their lives and the lives of their loved ones), rather than envisioning work without performance evaluation systems.

Workers frequently receive the corporate equivalent of “there is no alternative” when they raise the possibility that not all labor needs to be measured or that some valuable work defies quantification. After all, how can we know something is valuable if we can’t quantify it? Since we only value what we can measure, this circular reasoning becomes perfect. Fisher emphasized how even dissenting opinions are absorbed and turned into commodities by capitalism. In the same light, surveillance complaints are used as an excuse for “wellness programs” that only introduce additional metrics to monitor. This mentality even reduces “work-life balance” to just another metric to optimize. Performance metrics have been used to comprehensively individualize the mental health crisis in the workplace. It’s presented as your own inability to adjust rather than a systemic issue if you’re having trouble hitting your goals, feeling overburdened by continuous monitoring, or burning out from the never-ending demands of optimization. More metrics are always suggested as a solution: monitor your sleep patterns, gauge your stress levels, and count the number of minutes you spend meditating.

That supports what Fisher termed “business ontology” which is the idea that everything ought to be conducted like a business. Every human action can be turned into a business process with goals, benchmarks, and ROI calculations. Employee satisfaction ratings, engagement measures, and burnout risk assessments are just a few examples of how opposition to this system is quantified and marketed.

What we are witnessing is not merely a new phase of capitalism but, as economist Yanis Varoufakis argues, its transformation into something potentially worse: technofeudalism. The tech giants have created privatized digital fiefdoms where they don’t produce capital so much as extract rent from our every interaction. We’ve become what Varoufakis calls “cloud serfs,” building value for our digital overlords through our tweets (xheets?), posts, and clicks while lacking the class consciousness to recognize our own exploitation. The performance metrics that govern workers in this system aren’t tools of capitalist productivity but rather mechanisms of feudal control, ensuring compliance and extracting maximum value from an increasingly captive peasantry.

However, Fisher’s work also suggests possible escape routes. He maintained that acknowledging capitalism as a contingent system, as opposed to a natural rule, is the first step towards escaping its actuality. This also holds true for the regime of performance fetishization. These metrics can be deconstructed or drastically redesigned because they are human inventions, and they are relatively recent.

Reimagining productivity

The rot in our digital systems is not a flaw; it is the unavoidable result of institutions that define success by how effectively they can extract value. It’s as if we’ve created our entire organizational universe to answer the question, “How can we make both people and the planet maximally miserable while calling it optimization?”. The effects are already evident in the tech industry. An increasing skepticism over the direction of the industry, burnout, and union organizing activities (as a counter-reaction). Tech professionals are becoming aware that the very systems they are developing may use the same performance metrics to justify replacing them.

So here we are at the end of a crossroads. Our challenge as workers is to stay the course, or radically rethink what productivity means in an era of ecological crises and technological abundance. Our existing metrics are instruments intended to solve challenges of a different century. The big obstacle in front of us right now is how to control this abundance without causing ecological collapse. 

How do we strengthen the relationships that sustain us? How do we build feedback loops that help us understand our impact on the larger systems we’re part of? How do we design for regeneration rather than extraction? What may this actually look like? 

We can start by getting rid of work that isn’t needed. We should first determine which responsibilities are truly necessary to fulfill, rather than attempting to make every task more efficient. A large portion of what we now measure and optimize are bullshit jobs, work that exists solely to maintain employment in a system that requires continuous improvement. By getting rid of these pointless chores, we free up human energy for worthwhile endeavors.

We can prioritize caring over competition. Much of the most meaningful work that people do, such as caregiving, creative work, and emotional labor, cannot be quantified using conventional metrics. This requires a rethinking of productivity. We must find new ways to value and encourage this essential work rather than attempting to impose these activities into performance standards like they are work for industrial purposes.

We can build more circular systems. Productivity should be gauged by how well systems regenerate themselves rather than by linear output. This means placing an equal value on upkeep, restoration, and repair as well as new production.

We can prioritize investing on quality time. We should gauge how much time people have for creativity, connection, and care, rather than how many hours they work or how much they produce in an hour. A society that maximizes human thriving rather than labor extraction is one that is genuinely productive.

Some of these strategies are already being pioneered by unions and worker cooperatives, such as the Tech Workers Coalition. They provide room for more ecological and compassionate perspectives on productivity by giving employees a say in how their work is structured and evaluated. Of course technology may actually facilitate this shift if it is used wisely. This is a practical necessity rather than idealistic mentality. Both our well-being and the ecological systems on which we rely are being actively destroyed by the outdated concepts of productivity. We require new metrics (or do we?) for evaluating and assessing work that take into account the limitations of natural systems as well as the complexity of human needs.

The workplace of the future shouldn’t resemble a technofeudalist system. This is being demonstrated by companies that are experimenting with cooperative ownership, results-only work settings, and four-day workweeks. People naturally function well without complex measurement methods when you trust them and give them meaningful work that aligns with natural systems. The technology exists to build new forms of organization. The question isn’t whether reimagined organizations are possible – it’s whether we’ll create them before our current organizations finish consuming everything we value.

PS: Special thanks to Erin Casali, Lisa Baskett, Ron Bronson, and Dorian Freeman, for reviewing this essay, providing feedback, and challenging me in the notions presented here.