Creation by Deletion
“Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.”— Antoine de Saint-Exupéry, Terre des Hommes (1939)
Work Expands to Fill All Time Available
I recently had a personal insight about feeling like I constantly have too much going on. When I decide to take on a task or role, I didn’t notice there’s a hidden future tax on my time. Some things keep asking for more of your time and attention; others have a clear end point at which you no longer need to invest time or mental energy. That’s another framing on infinite vs. finite games. Infinite games are, well, infinite, and they tend to expand in size and importance over time. Finite games have a clear finish line at which point you can decide to stop or play again.
What sort of tasks keep demanding more of your time, attention, and energy? A family, a startup, Ironman training — these are things you should constantly be working on. They usually require a long-term commitment in order to do well.
On the other hand, finite tasks are, for example, automating your personal admin or delegating non-essential tasks. Or perhaps you commit to a creative project and know ahead of time what the end state will be.
This idea makes it clear why some things tend to just take all my free time away — and those infinite games compete with each other. Those big rocks compete with each other in such a way that I feel I’m always falling behind on one or many aspects at any point in time.
Bloat comes from unseen, future, and growing demands. Now the filter is whether doing something leads to doing even more in the future. You can play only so many infinite games that self-perpetuate before they compete with each other.
It’s freeing to know that, with that simple question, I’m protecting my future self from creeping, ever-growing demands that made me feel like I’m stuck in midlife molasses.
The Science of Bloat
In 2021, a team at the University of Virginia published a study in Nature that should have made every founder stop and stare. Across eight experiments with 1,153 participants, they found that people generate 8 additive suggestions for every 1 subtractive suggestion. Eight to one.
The Lego experiment made it visceral. Participants were given an unstable structure with a single wobbly column. Removing one block fixed it — for free. Adding blocks cost ten cents each. In the control condition, 59% of participants added unnecessary bricks rather than removing the one that was causing the problem. They paid money to make something worse because their brains couldn’t see the absence as a solution.
Your brain defaults to adding because subtraction is cognitively expensive. You can’t see, feel, or sense a void, so you have to work harder to grasp emptiness; your senses can’t help you. Removing instead of adding often feels very counterintuitive, awkward and uncomfortable. That’s our biases talking.
Removing, editing, curating are System 2 operations — slow, deliberate, effortful. Addition is System 1 — fast, automatic, comfortable. Your brain is basically choosing the drive-through over cooking. Every. Single. Time.
And even when you do think of subtracting, loss aversion pulls you back. Losses loom approximately twice as large as equivalent gains. In the famous endowment effect study, people given a mug demanded roughly twice the price to sell it as non-owners were willing to pay to buy the same mug — simply because it was already theirs. Cutting a feature feels like losing a limb, even when the feature adds no value.
A French replication confirmed the addition bias — but found that just telling people “remember you can also remove things” increased subtractive thinking 2.5x. The addition bias is default, not fate. You can override it. But only if you remember it’s running.
Parkinson’s Law
Northcote Parkinson wrote about the collective implications of the addition bias, initially in The Economist in November 1955. Between 1914 and 1928, the British Admiralty grew its officials from 2,000 to 3,569 — a 78% increase. During that same period, the Royal Navy shrank by a third in personnel and lost two-thirds of its ships. Fewer ships, fewer sailors, more bureaucrats. Parkinson found that organizations naturally bloat at 5-7% annually, regardless of how much actual work exists. He called it a law. It was really a diagnosis.
Bureaucratic entropy seems as predictable as thermodynamic entropy.
“Work expands so as to fill the time available for its completion.” - Parkin'son’s Law
The biology here is uncomfortable. In your body, programmed cell death — apoptosis — keeps tissues healthy by ensuring cell death exactly balances cell division. When that balance breaks — when cells refuse to die — the result is cancer. The cells aren’t malicious. They’re just growing. The problem is that nothing is stopping them.
Organizations are the same. University administrative staff grew 240% between 1975 and 2005 while faculty grew 50%. The ratchet effect means every hire, every tool, every policy creates dependencies that resist deletion. Researchers call it organizational debt — the slow accumulation of zombie processes and legacy policies nobody remembers creating.
In last week’s “Entropy-neurship“ post, I talked about channeling entropy. This is a unique role founders can play to help shape better systems: companies, public institutions, etc.
Why? As we’ve seen, people are biased toward adding and doing more — and need a reminder that they can, and probably should, also be removing. Employees need to perform and prove their value, usually by taking decisive actions. Doing nothing as a strategy tends to get less social approval — and is harder to defend when things go wrong. (Were you lazy? How do you measure the impact of doing nothing?)
The meta-evidence is clear. An integrative review of empirical studies found that constraints consistently benefit innovation, with a sweet spot of moderate constraint (Acar et al., 2019). And founder CEOs produce 31% more citation-weighted patents than professional CEOs; when a founder departs, innovation drops by roughly 44% (Lee et al., 2016).
Founders are the ones incentivized to get it right — to make the right decisions, to design the right solutions. And this is why they are typically the only people inside a company who are able and willing to design by reduction — and question whether something should exist at all.
The Scar Tissue
Nassim Taleb wrote that “knowledge grows by subtraction much more than by addition.” Knowing what NOT to do is more durable than knowing what to do. But that knowledge comes from scar tissue — from the deletions that went wrong.
The secret is subtraction. Not once, but always. Constantly.
Strategic subtraction is about design tradeoffs. To delete well, be clear on what is being removed and what the impact could be:
“If I removed this, what would break?”
“Am I cutting cost or cutting capability?”
“Am I optimizing for efficiency or for discovery?”
Designing by subtraction is clearly not the same as cost cutting. In fact, one is about efficiency, the other is about innovation (very inefficient). The key is that removing should set you up for a better future — not destroy your potential. With that frame, the right type of strategic subtraction is a form of investment.
As we’ve seen last week, the right way to harness entropy is to channel it, not fight it — so the right way to subtract has to be in harmony with a broader whole: removing to reallocate energy better.
“We shape clay into a pot, but it is the emptiness inside that makes it useful.
We cut doors and windows for a room; but it is the empty space that makes the room livable.
We work with being, but non-being is what we use.”
— Lao Tzu, Tao Te Ching, Chapter 11
The Japanese have a word for this: Ma. The intentional negative space. The silence between notes that gives the melody meaning. The empty wall that makes the painting visible.
Every wisdom tradition found this independently. Taoism teaches that non-being is what we use. Buddhism’s sunyata holds that emptiness is the undifferentiation from which all things arise. Michelangelo defined sculpture as “that which is fashioned by the effort of cutting away” — the artist reveals, not creates.
Perhaps the founder’s most important work is invisible. The features you didn’t build. The hires you didn’t make. The markets you didn’t enter. The meetings you didn’t hold. Nobody sees the absence. Nobody applauds the product you killed before it shipped.
But the absence is what lets everything else exist and flourish.
Happy Sunday ✌🏼🌖
Sources
Adams, G. S., et al. (2021). People systematically overlook subtractive changes. Nature, 592(7853), 258-261.
Fillon, A., et al. (2025). The Overlooking of Subtractive Changes: Replication and Extension. Journal of Creative Behavior, 59(1).
Kahneman, D., & Tversky, A. (1979). Prospect Theory. Econometrica, 47(2), 263-291.
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental tests of the endowment effect. Journal of Political Economy, 98(6).
Kerr, J. F., Wyllie, A. H., & Currie, A. R. (1972). Apoptosis. British Journal of Cancer, 26(4).
Matejka, M., et al. (2024). The Ratchet Effect. Management Science, 70(1).
Kocak, S. A., et al. (2024). Organizational debt. PLOS ONE, 19(11).
Ginsberg, B. The Fall of the Faculty. Johns Hopkins University Press.
Lee, J. M., et al. (2016). Are Founder CEOs Better Innovators? S&P 500 Firms.
Acar, O. A., et al. (2019). Creativity and innovation under constraints. Journal of Management, 45(1).
Iyengar, S. S., & Lepper, M. R. (2000). When choice is demotivating. JPSP, 79(6).
Govindarajan, V., et al. (2025). Strategic Subtraction. Harvard Business Review.
Larsen, B., & Luna, B. (2024). Cognitive control during development. Journal of Neuroscience, 44(26).

That was a fantastic read. I wasn't aware of System 1 vs System 2 operations, but the framework makes a lot of sense. Now I definitely need to read Thinking, Fast and Slow (it's been on my list for years, but other infinite games have kept me away 😎).