The smell of ozone and burnt hydraulic fluid is a permanent resident in Camille D.’s nostrils. She is currently suspended 118 feet above the lobby of a mid-rise office building that smells faintly of stale coffee and broken promises. Her flashlight flickers, catching the glint of a governor cable that looks like it has been chewed by a very determined metallic rat. Camille D., an elevator inspector with 18 years of grime under her fingernails, doesn’t panic. She just sighs. This isn’t a mechanical failure yet; it is a failure of intent. The elevator works-barely-only because a janitor named Hector has been applying a specific brand of unauthorized lubricant to the guide rails every 8 days for the last 18 months. Hector isn’t a technician. He’s just a guy who hates the sound of the screeching car, and in doing so, he has accidentally become the only reason the building’s vertical transit system hasn’t turned into a 4,008-pound plummeting coffin.

I lost an argument yesterday. It was one of those exhausting sessions where 8 people sit in a glass-walled room and debate the ‘optics’ of a failure. I was right-I know I was right-about the 108th row of a spreadsheet that proved our projected timeline was a fantasy. I pointed out that the only reason we hit our last 28 milestones was that a junior developer was working until 2:08 AM every night to fix bugs that shouldn’t have existed in the first place. My boss looked at me and said that we should ‘celebrate the dedication’ of that developer. I told him we weren’t celebrating dedication; we were subsidizing bad architecture with a human being’s sleep cycle. I lost the argument because, in the eyes of the institution, a problem that is being ‘handled’ isn’t a problem at all. It’s a success story.

The “Yes, And” of Dysfunction

Camille D. taps the frayed cable with her wrench. She knows that if she writes this up as a critical failure, the building shuts down. If she doesn’t, she’s relying on Hector and his unauthorized grease. This is the ‘yes, and’ of systemic dysfunction. Yes, the system is broken, and yes, we will find a way to make it look like it isn’t. We externalize the fragility of our organizations onto the most conscientious people we can find. We find the person who cares too much to let the ball drop, and then we load them up with every ball the system is designed to drop. We call them ‘rockstars’ or ‘MVPs’ because those titles are cheaper than actually fixing the workflow. It is a form of gaslighting where the reward for being reliable is more weight.

Built-in Dependability vs. Human Patchwork

When we look at the way products are crafted in a world obsessed with these ‘heroic’ workarounds, the contrast is jarring. In a truly robust system, quality isn’t a frantic last-minute correction; it’s a foundational requirement. This is something I’ve been thinking about in terms of the way we treat our dependencies-not just in business, but in life. For instance, when we look at something like Meat For Dogs, there is a distinct lack of this ‘duct-tape’ mentality. There is no one in the background trying to fix a fundamental flaw in the ingredients with clever marketing or hidden compensations. The dependability is baked into the design, not added as a desperate layer of human goodwill after the fact. It’s the difference between an elevator that runs because it was built to run and one that runs because Hector showed up with a grease gun.

There are approximately 88 ways to hide a failing process. You can hide it in ‘culture,’ you can hide it in ‘agile methodology,’ or you can hide it in the ‘extra mile’ that you expect every employee to run. But the extra mile is supposed to be an exception, not the standard operating procedure. When you make the extra mile the baseline, you are no longer running a company; you are running a marathon of attrition. Camille D. remembers a job 8 years ago where she was told to ignore a faulty secondary brake because the building manager promised they had a ‘monitoring system’ in place. That monitoring system turned out to be a security guard who was told to listen for a specific clicking sound. The guard eventually quit, the clicking went unheard, and the repair ended up costing $8,888 more than it would have if they had just fixed the damn brake in the first place.

Addicted to the “Anita’s”

We have become addicted to the ‘Anita’s’ of the world. We love the person who stays late. We love the person who ‘just gets it done.’ But we never ask why it requires someone staying late or why it wasn’t ‘done’ by the system we spent 48 weeks and several hundred thousand dollars implementing. By praising the individual’s ability to overcome the system, we give the system a free pass to remain broken. It’s a parasitic relationship. The system drains the goodwill of the person until they burn out, and then the system looks for a new host. It’s a cycle that repeats every 18 to 28 months in most corporate environments.

I think back to that argument I lost. I realized later that the reason I lost wasn’t because my data was wrong, but because my data was an inconvenience to the narrative of ‘dedication.’ If I am right, then the leadership has to do the hard work of redesigning the process. If they are right, they just have to buy the junior developer a $18 pizza and a ‘thank you’ card. The choice, for them, is a matter of simple, albeit cruel, math. They are trading 8 hours of a developer’s sanity for 80 hours of their own comfort. It is a trade they are willing to make every single time because they aren’t the ones paying the interest on that debt.

The Violent Honesty of Failure

Camille D. eventually steps out of the elevator shaft. She decides to red-tag the car. She knows the building manager will scream. She knows they will bring up the ‘resilience’ of the building and how they’ve never had an accident in 38 years. But she also knows that Hector’s grease is a lie. It’s a beautiful, kind-hearted lie, but it’s a lie nonetheless. She writes the report with a precision that borders on the aggressive, documenting all 18 points of failure that the human element had been masking. She feels a strange sense of relief, the same kind I felt when I finally stopped trying to ‘fix’ the spreadsheet for my boss and let the errors populate as they were meant to. There is a certain violent honesty in letting a bad process fail.

We often talk about ‘systemic failure’ as if it’s a sudden event-a crash, a collapse, a catastrophe. But systemic failure is usually a slow, quiet process that is being held at bay by a few exhausted people. It’s the nurse who works a double shift because the scheduling software is a joke. It’s the teacher who buys supplies because the district budget is a black hole. It’s the programmer who patches the 8,008th line of code at 3:00 AM because the original architect didn’t believe in documentation. These people are the heroes of our society, but they are also the enablers of our dysfunction. By being so dependable, they allow the people in charge to remain deluded about the quality of the systems they oversee.

Building for Durability, Not Attrition

If we want to build things that actually last-things that are truly dependable-we have to stop relying on the ‘extra mile.’ We have to build systems that are boringly efficient, processes that work even when the most talented person in the room is having a bad day. We have to stop externalizing the cost of our bad decisions onto the goodwill of others. Because eventually, Anita will retire. Hector will find a better job. Camille D. will put down her wrench. And when that happens, the system won’t just fail; it will reveal that it was never really there to begin with.

What would happen if you stopped compensating for the broken things around you for just 8 hours? Who would be the first to notice that the machine was never actually running?