In 1989, Stephen Covey published The 7 Habits of Highly Effective People. It sold 40 million copies and became one of those books that people reference without having read. I was one of those people until recently.
The seven habits are fine. Be proactive. Put first things first. Sharpen the saw. They live on posters in dentist offices for a reason — they're true in that sturdy, obvious way that makes you nod without changing anything.
But buried inside Habit 2 — "Begin with the End in Mind" — is an exercise that actually stuck.
Covey asks you to imagine your own funeral. Four speakers: a family member, a friend, a colleague, and someone from your community. What do you want each of them to say about you?
Not what they would say. What you want them to say. The gap between those two things is your work.
It's a brutal exercise because it forces clarity. You can't hide behind "I want to be successful" when you're imagining your daughter at a podium, describing what kind of parent you were. Success becomes concrete: Did I show up? Did I listen? Did I make her feel seen?
Covey's point is that all things are created twice — first mentally, then physically. The funeral test is the mental creation. You deciding what "winning at life" actually means before the world decides for you.
The company I work for has a document called "The Codex." It's not a style guide or an employee handbook. It's a 6,000-word operating system for how to think — cognitive clarity, strategic positioning, decision-making frameworks, philosophical commitments.
When I first encountered it, I thought: this is intense. Then I realized it's just the corporate funeral test.
Most companies put values on a wall — integrity, innovation, excellence — words so generic they could apply to a hospital, a hedge fund, or a hot dog stand. The Codex is different. It doesn't say "we value productivity." It says: At the end of each week, ask: Did I change reality for a customer, or did I only change Notion?
That's specific enough to be falsifiable. You could actually check whether someone is living by it. And that specificity is everything — generic values don't shape behavior. Articulated mental models do.
The Codex answers the corporate funeral question: if this company died tomorrow, what would we want people to say it stood for?
I've been trying to build something like this for myself. Not a manifesto — more like a set of diagnostic questions I return to when I suspect I'm drifting.
In software, a test harness is a framework that verifies your code behaves correctly. You define expected outputs, run the code, and check whether reality matches intention. The funeral test is one of these: "What do I want people to say about me?" is the expected output. My daily actions are the code. The gap between them is the failing test.
But I can get more specific.
It's 7pm on a random Tuesday. No special occasion. How am I spending my time? If the answer doesn't align with what I'd want my funeral speakers to describe, something's off.
What advice do I give other people that I don't follow myself? That's a failing test. Either update the advice or update the behavior.
Where did my discretionary money go last month? Money is crystallized priorities. It doesn't lie the way aspirations do.
These aren't meant to induce guilt. A failing test isn't a moral failure — it's information. It tells you where intention and action have diverged, so you can recalibrate.
And sometimes the tests are flaky. In software, a flaky test fails intermittently — not because the code is broken, but because the conditions weren't right. You were tired. The environment was off. Something external interfered. Run it again later and it passes.
I have plenty of flaky tests. I spiral about whether I tipped enough at dinner. I fixate on something stupid I said in passing. There are days I really fumble — not because I'm a bad person, but because I was depleted, or distracted, or just having a contracted day.
You're allowed to have those days. Days where you're not the most generous person in the room. Where you don't pass the Tuesday Night Test because you needed to stare at a wall.
The goal isn't a perfect pass rate. It's noticing. Catching the pattern over time. Distinguishing between I failed this test once because I was exhausted and I've been failing this test for six months and something needs to change.
Self-compassion isn't the absence of standards. It's knowing when a failure is signal and when it's noise.
I have a three-year-old daughter.
She won't remember most of what's happening right now. The specific toys, the exact meals, the individual bedtimes — they'll blur into a general feeling. What she'll carry is the texture of her childhood. Whether she felt safe. Whether she felt seen. Whether home was a place of joy or tension.
So I run her version of the test.
Twenty years from now, when she thinks about growing up, what will the highlight reel look like? Not the Pinterest moments — she won't remember those. The recurring patterns. The feel.
Kids learn more from what you do than what you say. When I'm frustrated, do I blame or take responsibility? When I'm wrong, do I apologize? When I'm stressed, do I reach for my phone or for presence?
"My mom was always on her phone" is a story someone's kid will tell. So is "My mom put her phone away when I talked to her." Both are true for someone. The question is which one is true for mine — not on the days I'm trying, but on the random Tuesdays when I'm tired and she's whining and my phone is right there.
That's where the real test runs. Not in the aspirations. In the defaults.
We're entering an era where AI agents will represent us. They'll answer emails, manage calendars, maintain relationships while we're busy or asleep. The question of how those agents know who we are is just the funeral test in new clothes.
Right now, they infer it from your exhaust data. But there's another approach — tell them. Write it down. Declare who you are instead of letting the pattern-matching decide.
That's a practical problem worth solving, and I've written about it separately. But the deeper point isn't about AI at all.
There's a nature you discover — your wiring, the things that were always there but needed words. And there's a character you build — your values, your habits, how you respond to what life throws at you. Authenticity is the whole picture. You find the raw materials. Then you decide what to make with them.
The question isn't whether you'll be defined. You will be — by algorithms, by colleagues, by your kids, by the patterns you repeat without thinking. The question is whether you'll do any of the defining yourself.
Begin with the end in mind. Then check your work.