The Glitch in the Hull
I didn’t expect the brain freeze to hit that hard. One minute I was digging into a stolen bowl of vanilla bean in the galley of the USS Memphis, and the next, a shard of ice had pierced the roof of my mouth and was currently performing a lobotomy on my frontal lobe. I stood there, leaning my forehead against the cool, damp metal of the bulkhead, waiting for the world to stop vibrating. It was the 16th time this month I’d retreated to the walk-in freezer just to hear myself think. Above me, on the other side of the hatch, the leadership was having another meeting. I could hear the muffled drone of the Executive Officer’s voice. They were talking about the ‘Hydra System’-a piece of sonar software that had been glitching since the 46th day of our deployment.
We all knew it was garbage. The tech specs were bloated, the UI was a nightmare of 1990s-era buttons, and it had a nasty habit of misidentifying whale songs as hostile torpedoes. We should have scrapped it. We should have gone back to the legacy system that actually worked. But we didn’t. We couldn’t. Because the Hydra System wasn’t just a piece of software; it was a career path. A high-ranking Admiral had staked his next star on it, and back here on the sub, our CO wasn’t about to be the one to tell him his baby was ugly. So, we spent 86 hours a week trying to patch a sinking ship with digital duct tape.
“
The system wasn’t dead because it failed the metrics; it was kept alive because it fueled a promotion. That’s not engineering; that’s bureaucracy in high-pressure steel.
Project Phoenix: Animated by Ego
It’s the quarterly review for ‘Project Phoenix.’ The air in the conference room is that specific brand of recycled oxygen that smells like carpet cleaner and desperation. The metrics on the screen are a sea of red-downward trends that look like the EKG of a dying patient. The team is slumped in their ergonomic chairs, eyes glazed over. We’ve been at this for 16 months. The market has moved on, the competitors have launched three superior versions of our core feature, and our primary beta tester just sent an email that was mostly just the word ‘HELP’ repeated 26 times.
“
The VP of Product stands up. He doesn’t look at the red lines. He looks at the horizon. He’s wearing a watch that costs more than my first car-somewhere in the neighborhood of $5506, if I had to guess. He clears his throat and says, ‘The data is challenging, sure. But we’ve invested too much to pivot now. We just need to double down. We need more velocity. More synergy. More… Phoenix.’
I feel that brain freeze coming back, even though there’s no ice cream in sight. It’s a cognitive dissonance so sharp it physically hurts. We are witnessing the birth of a zombie. A project that is operationally dead but remains animated by the sheer force of political will. We usually talk about the ‘sunk cost fallacy’ as if it’s a simple math error. We think, ‘Oh, I’ve already spent $66 on this bad movie ticket, I might as well stay and watch the ending.’
When a project is failing, the logic shifts from ‘Is this creating value?’ to ‘Who gets blamed if we stop?’ For the VP, killing Project Phoenix isn’t an act of strategic wisdom. It’s a public admission of failure. It’s a stain on a 26-year career of ‘winning.’ As long as the project exists, there is a 6 percent chance that a miracle might happen. If he kills it, the probability of failure becomes 106 percent. So, the zombie walks.
The Rot Beneath the Scented Spray
I remember Miles M.-L., the cook I mentioned earlier. That’s me, by the way. I’ve spent more time in the bowels of ships than in boardroom meetings, but the smell of a failing project is exactly like the smell of a grease trap that hasn’t been cleaned in 76 days. You try to ignore it. You mask it with scented sprays. You tell yourself it’s just ‘character.’ But eventually, the rot becomes the environment.
When we refuse to kill a bad idea, we aren’t just wasting money. That’s the smallest part of the tragedy. The real cost is the morale of the people forced to perform the charade. There is nothing more soul-crushing for an engineer than being told to optimize a feature that nobody wants. It’s a form of professional gaslighting. You know the work is meaningless, but you are required to show up to the stand-up meeting and talk about your ‘wins’ with a straight face.
“They leave behind the people who are comfortable with the charade, the ones who have mastered the art of looking busy while a ship sinks at a rate of 6 knots.”
This creates a culture of pervasive hypocrisy. If the leadership is lying about the viability of the project, what else are they lying about? The trust evaporates. The best people-the ones who actually care about making things that matter-are the first to leave. They don’t want to be necromancers; they want to be builders.
The Paprika Effect: Trying to Save Rotten Beef
47 Hours of Paprika
Captain Relieved
I once spent 46 hours straight trying to salvage a batch of beef bourguignon that had been tainted by a refrigeration failure. I added spices, I reduced the sauce, I tried every trick in the book. I didn’t want to tell the Captain I’d lost the main course for the dignitary dinner. But at hour 47, I tasted it. It was still rotten. No amount of paprika can hide the taste of decay. I had to throw it all out and serve grilled cheese sandwiches. And you know what? The Captain wasn’t mad. He was relieved I didn’t poison the Admiral.
In the world of product development, we rarely have the courage of the grilled cheese sandwich. We keep adding paprika to the rotten beef. We write 66-page post-mortems that don’t actually admit the project died months ago. We create ‘Project Recovery Task Forces’ that consist of the same 16 people who broke the project in the first place.
Burying the Dead to Fund the Living
To break this cycle, we have to decouple reputation from results. We have to make it safe to say, ‘This was a good hypothesis, but the data says we were wrong.’ True engineering discipline isn’t about the ability to build anything; it’s about the discipline to stop building the wrong thing. Companies that thrive, like
Lando, understand that the ‘kill switch’ is just as important as the ‘launch’ button. It’s about the integrity of the process, not the ego of the sponsor.
Every time we let a zombie project live, we are starving a good idea of oxygen. There is only so much budget, so much time, and so much emotional energy in a room. For every Phoenix that is kept on life support, there are 6 brilliant, small, innovative ideas that never get funded. They die in the shadows because we are too busy pretending the corpse in the corner is just taking a nap.
I think back to that meeting on the sub. I eventually stopped leaning against the bulkhead and went back to the galley. I didn’t fix the Hydra System. I couldn’t. But I did stop pretending I liked it. When the XO came down later and asked how the sonar was holding up, I didn’t give him the company line. I told him it was a bucket of bolts that couldn’t tell a humpback from a Harpoon. He stared at me for 6 seconds, then sighed and took a bite of my grilled cheese.
– Miles M.-L.
‘I know,’ he whispered. ‘But we have to keep the Admiral happy.’ That’s the trap. We are all trying to keep some Admiral happy, even when the Admiral is 1006 miles away and doesn’t have to live with the consequences of his bad decisions. We perform the dance. We update the JIRA tickets. We hold the retrospectives where we talk about ‘learnings’ instead of ‘failures.’
Burying the Dead
But the brain freeze always returns. It’s the physiological reminder that something is wrong. That we are consuming something too cold, too fast, too artificial. If you are sitting in a meeting right now, looking at a slide deck for a project that you know is a zombie, do yourself a favor. Stop adding the paprika. Admit the beef is rotten. It might be a messy conversation, and it might even cost someone a promotion, but it’s the only way to clear the air.
Every zombie project is just a monument to someone’s inability to say ‘I was wrong’ in public.
I’ve seen projects that lasted 256 days past their expiration date. I’ve seen teams spend $856 per hour on consultants to tell them what they already knew. It’s a waste of human potential that makes my heart ache more than any ice cream headache ever could. We have to be better than our sunk costs. We have to be brave enough to bury the dead so the living can finally get some work done.
The courage to stop is the highest form of product discipline.
Comments are closed