There’s less friction in pursuing dogma than embracing multiplicity. Naturally, the vibrant colors of our world fade into black and white. This is the act of optimization. It’s an unfortunate one, because to optimize is to lose optionality.
Optionality is not only about having a plurality of options, but also fostering asymmetric upside; your loss is limited by the investment you make in an option—whether through capital, time, or effort—but each one can potentially bring outsized returns.
I’m sensitive to optimization, because a large part of my life philosophy is simply about creating and preserving optionality. This is why the concept of learning environments is so important: there are environments where inference will lead you astray, so times exist when you must forget everything you’ve ‘learned’. This sort of meta-situational awareness is key in giving yourself permission to press the eject button when optimization has gone too far.
More often than not: people don’t realize this and continue answering the wrong questions. It’s a pattern my first two professional experiences established for me.
Herbert Simon puts words to this pattern in his acceptance speech for the Nobel Prize in Economics, where he cogently said:
Decision makers can satisfice either by finding optimum solutions for a simplified world, or by finding satisfactory solutions for a more realistic world.
I always strive for the latter—or at least delude myself into thinking so. You find the former in the mediated realities that wrap around us; a threadbare, patchwork curtain that can be ripped apart if gripped tightly. More often than not, its incompetence reveals on its own. The wool pulled over your eyes is soft, but itchy.
I first observed this at a summer restaurant job, the type where a customer’s order is constructed step-by-step as they walk down the counter.
My general manager cared about only one thing: throughput. This was the number of customers who passed the cash register per hour. As a leading indicator of our revenue, team members were expected to prioritize it at all costs.
The led to stunted customer interactions, such as minimal eye contact, blitzkreig-style questioning, and an easily sparked impatience when—god forbid—someone began pondering the menu. The price of these social transactions was undoubtedly paid through future lost earnings.
This short-sighted management is clearly inadequate, if not wholly broken. It’s the type of error that provides fodder for cubicle caricature like Office Space or The Office. However, our societal reaction was to double-down on optimization, with the result being what you find in most workplaces today. Instead of uprooting dogmatism, we responded by saying you needed more metrics to measure—or as one popular organizational theorist has put it, to measure what matters.
In response to the limits of conformity to a single rule, we established the conformity to many. Adding more metrics was supposed to get us closer to the truth. However, they just added complexity to already nebulous settings. To make sense of it all, as well as to act, people draw on innate tendencies—tribal behavior—enabling the emergence of power silos and information asymmetries.
My summer after the restaurant, which I spent planning the production of medical devices at a local factory, cemented this point for me. Individual departments—Supply Chain, Operations, Safety, Quality, the operators themselves—engaged in a group juggling act to maximize the metrics they were responsible for. If Operations wanted to speed up production, Safety would be concerned about increased injuries that could occur from rushed soldering. If Quality wanted to implement a new test for defective parts, Supply Chain would rush in, and so on.
It was territorial, zero-sum behavior under the guise of coordination. In all the numbers improved that summer conducting ‘process improvement’—just take a look on my LinkedIn—I never felt that other teams beside my own benefited from or appreciated the changes. This is due to the concept of ownership; the metrics I increased weren’t the ones they were responsible for. If anything, other departments actively hindered progress in favor of their own plans, only slinking away once a higher power—such as the plant manager—was called.
This type of thinking is the foundation of modern meritocracy and accountability—the extent to which somebody can meet their numbers. Alongside how this creates in-groups and out-groups, like in my factory setting, it also fails because goals are intrinsically pathological and empower you only to the extent you follow them. We create sub-goals for our goals and so on: team goals under our organizational ones, individual goals under our team ones, etc. The deeper you go into the rabbit hole—and there are many we dig—the more disconnected you become.
In most cases, employees of an organization don’t want to question the goals they’re given; they’ll follow them even when these goals stop mapping to reality or the outcomes they wish to achieve. No one gets fired for following their job description. The results are lifeless, bureaucratic organizations following outdated scripts. They grow, but only in a cancerous sense: mindless reinforcement and perpetuation.
True power does not lie in those who create, but those who remove.
Organizations are pathological—many modern ones exist as a bundle of goals—and will resist disruption. This is inherently clear in their scaffolding for decision-making. In a hierarchical company, the power to announce a goal as dead or irrelevant will often lie in the founder or CEO. This is also likely the last person in the company to recognize a goal as irrelevant; the whispers of frontline workers occur far before any palpable corporate dysfunction. This is also visible in mainstream organizational programming. Corporate innovation programs incentivize employees to implement new ideas, not tear down old ones.
We’d do a lot better to question the goals we’re given—both personally and professionally—instead of immediately seeking to maximize. Often, it will be the case that we need to remove them. When I facilitated goal-setting for a startup incubator, the work that teams found most valuable was the challenging of assumptions behind their goals.
So far, this has been a case for intentionally resisting optimization, but often it’s also enough to simply allow entropy to play.
In my freshman year of college, I set out to pursue my dream of becoming a doctor. My mediocre chemistry and math grades quickly convinced me otherwise. The department I believed I was made for—computer science—rejected my application, and I settled for a ‘lesser’ major. It would prove to be all I needed. However, to conclude my time as a student, neither of my internships extended a return offer.
These serendipitous points of failure did not only preserve my optionality, but developed it. Three years ago, I wouldn’t have been able to predict anything close to my current interests, relationships, setting, and career. These were never supposed to be options on the table. What was written in my script—being a doctor or software engineer, working at tried-and-true companies—now seems to have had the wrong person in mind. I’m glad I didn’t hold on to those pursuits too tightly, gritting my teeth to persevere or try again. I might have succeeded.