Watch

Listen

“Research is formalized curiosity. It is poking and prying with a purpose.”

Zora Neale Hurston

Two years of thinking never told me why I struggle to disconnect from electronics by 10pm. One week of experiments did.

Each night I ask myself one question: what’s my limiting factor? For two years the answer was mostly the same. Disconnect by ten. Most nights I missed it. Sitting with the question told me nothing.

I once wrote software for a living. When a program breaks and I can’t reason out the cause, I stop staring at the code. I drop a print statement in, run it again and let the machine show me what I can’t see. So I tried it on myself.

That 10pm rule is an anti-habit: I decided it once so I stop deciding it each night. It kept failing. So a week ago I dropped in a second rule as my print statement: finish work by 6pm, to see what it moved.

The two rules moved together almost perfectly. Each night work ran past six, I stayed on my screens past ten. The experiment found what two years of thinking missed.

I was failing at ten because I was failing at six, with not enough hours left to decompress before bed. The buffer was the thing I lacked.

You cannot reason your way to a hidden cause. The bug hides until you instrument the system.

In 1847, a Vienna doctor named Ignaz Semmelweis watched mothers die of fever on his ward at several times the rate of the ward next door, the one run by midwives. The medicine of the day blamed bad air. Semmelweis read the gap between the two wards instead and found the one thing that set them apart: his doctors came to the delivery room straight from the autopsy table. He made them wash their hands in chlorinated lime. Deaths on his ward fell by nearly ninety percent. The correlation found the cause the theory missed.

Seven years later, cholera was killing Soho in London. The authorities blamed poisonous air. A physician named John Snow plotted each death on a map. The dots clustered around a single water pump on Broad Street. His map found the source that reasoning alone missed.

Even the smartest people still need to run experiments to find the cause. So do you. In 1986, investigating why the Challenger shuttle broke apart, Richard Feynman skipped the argument. On live television he clamped a slice of O-ring rubber, dropped it in a glass of ice water and pulled it out stiff. Cold rubber stays stiff instead of springing back. Thirty seconds of experiment settled what months of memos only circled.

All three settled it by running tests.

Change one thing on purpose. Watch what moves with it.

I found the bug behind my late nights. Unfortunately, I haven’t found a single experiment that cracks every problem yet. But cracking a problem is a function of attempts: run enough experiments and the cause surfaces. Run experiments at a high enough rate and you can engineer success.

🔦

What experiment would answer a question you’ve been stuck on for years?

Watch

Listen

“Any intelligent fool can make things bigger, more complex and more violent. It takes a touch of genius and a lot of courage to move in the opposite direction.”

E. F. Schumacher, 1973

Sears was the biggest retailer in the world. Then it got stuck in the middle of the market. Walmart owned the low end on price. Nordstrom owned the high end. Sears, squeezed between them, died in the gap.

Basketball has the same middle. The mid-range jumper is the least efficient shot in the game. For a decade, NBA teams deleted the shot on purpose. They kept the layup and the three, cut everything between, then scored more efficiently than the game had ever seen.

Money works the same way. Nassim Taleb calls it the barbell. Put most of your savings somewhere safe. Put a sliver on the wildest bet you can stomach. Skip the middle.

• The safe pile survives a crash.
• The wild sliver catches the upside.
• The middle takes the fall.

The middle looks safe, so most money sits there and a crash hits it hardest.

My jogging habit runs on the same shape. A floor and a bar. Three minutes is the floor, short enough to survive my worst day. Eight is the bar, high enough to stretch my best.

A single target would try to do both and do neither. So I put the habit on a range instead of one number.

For a while five minutes sat in the middle of the range, the number I was meant to aim at. Five minutes survived no bad day and stretched no good one. I cut it.

Dieter Rams built a career on this: less, but better.

The middle number felt like discipline. It was only dangerous decoration.

🏀

What could you cut from a system you run to make it better?

Watch

Listen

“Grant me the serenity to accept the things I cannot change, the courage to change the things I can, and the wisdom to know the difference.”

Reinhold Niebuhr

I broke my own meditation more times than I can count. Same mistake, fifteen years running.

The practice was working. So I poked at it. Maybe I could shorten the 2 hours. Maybe swap in something faster.

Then it fell apart.

• My decisions got worse
• Anxiety crept back
• Sleep degraded

The gap between the change and the fallout ran a week or two. Long enough that I blamed something else each time.

The 2 hours are a wall I stopped trying to move. I pay them in full or I pay in anxiety, sleep and bad calls. The price holds no matter what I try. So I leave it alone.

Some walls have a door. I learned to spot them from a video game.

In a documentary called The Thinking Game, a DeepMind program teaches itself Atari Breakout. Paddle at the bottom, ball rising, a wall of bricks up top. For a while it does the obvious thing and chips through the wall one brick at a time.

Then it stops. The program digs a narrow channel up one side, slips the ball behind the bricks and lets it ricochet along the ceiling. The wall destroys itself. The paddle barely moves.

That move sat in the game the whole time. Players miss it because they keep sliding the paddle out of habit. I named it tunneling: a trade-off you thought was mandatory turning out optional all along.

My own tunnel was hiding in the work week. Everyone accepts that running a business eats most of your days. I cut from two operations days a week to one and braced for the slowdown.

Email did slow. Support got faster. I switched support from something I check to something that reaches me the second it lands.

Half the admin time, quicker replies. The wall had a door.

Doors turn up everywhere once you look.

Dick Fosbury cleared the high-jump bar going over backwards. Everyone else sprinted up and straddled it face-down. The rule said clear the bar.

How was left to him. He took gold in 1968 and each jumper alive copies him now.

Malcom McLean stared at a loaded ship and saw weeks of dockworkers hauling crates by hand. The shipping world called that the cost of trade.

He put the whole truck trailer on the boat instead. Loading time and theft collapsed. The steel box he standardized carries most of what you own.

Some walls stay walls. A rocket burns fuel to carry its fuel and the math wins each time.

Sleep is the same. Trade hours of it for work and the bill lands in your body instead of your calendar. Physics charges the same however clever you are.

So the skill is telling a door from a wall. I treated a wall like a door for fifteen years, poking at a practice that broke each time I touched it. I treated a door like a wall for years, paying for a work week that had a shortcut all along.

If effort changes the price, it was a convention. If nothing changes the price, it’s physics.

Accept the real walls. Tunnel the fake ones. Most walls are real and that’s exactly what makes the rare door worth the search.

🚪

What’s one cost you accept that you could push on instead of paying?

Watch

Listen

“It will not always be summer, build barns.”

Hesiod

Some nights I’d binge until 3am, sometimes 5, then wake up ashamed. My phone, a chess app, YouTube. The harder the day, the worse it got.

For a long time I read those nights as a discipline failure. I threw more willpower at them. The binges kept coming.

The real problem was simpler: some days demanded more than I could give. By night I had nothing left, so I binged.

So the obvious fix was to do less. Run fewer things.

I never did.

I still run five products.

The fix was habit banking: on a strong day I do some of tomorrow’s work early and bank the surplus. When a hard day comes, it spends that reserve instead of breaking the run.

No single day runs me into the ground anymore.

The binges stopped.

The fix for too much wasn’t doing less. The fix was building a buffer.

🕒

What could you get ahead on today so a harder day costs you less?

Watch

Listen

“A map is not the territory it represents, but, if correct, it has a similar structure to the territory, which accounts for its usefulness.”

Alfred Korzybski

I sent an email about putting my habits on a range instead of a single number. My meditation runs 55, 89 and 144 minutes, a Fibonacci sequence, more on a smooth morning and less on a tight one.

A reader hit reply. “What’s so special about the Fibonacci numbers? Why not just use 15, 30, 60, 90 or something like that?”

Sure, 15, 30, 60, 90 could work. My range is a model. So is that set. Some fit better than others. The reader is treating my range like a law of how meditation works.

Ride a subway and you’re trusting a model. Harry Beck drew the London Underground map in 1931 with straight lines and even spacing, the stations placed for legibility instead of location. The map is a distortion.

• The distances are off.
• The angles are invented.
• The real city underground looks entirely different.

Riders use it anyway. It answers the only question you have down there: which train, which direction, where to change. Beck traded accuracy for something you could read at a glance.

Newton didn’t know he was wrong. For two centuries his laws looked like the truth, until Einstein showed that mass bends space and Newton’s version misses it.

We flew to the moon on Newton’s model anyway. The math is simpler, faster to run and close enough to land a spacecraft. A wrong model still did the job.

Check tomorrow’s weather and you lean on another model. The weather is too chaotic to pin down. Forecasts often miss. You check it before you leave the house anyway, because a rough read on the weather beats walking out the door blind.

Phil Libin, who built Evernote, uses a model of his own at company scale. He calls it the rule of 3 and 10: three people become ten, then thirty, then a hundred. At each jump the way the company runs breaks and gets rebuilt. He says that the numbers aren’t a law of physics but a way to see the breaks coming.

The subway map, Newton’s laws, the forecast, Libin’s rule: each is useful without being true. Derek Sivers wrote a short book on the idea and named it Useful Not True. A belief doesn’t have to be accurate to be worth holding. It just has to pay you back more than the accuracy it costs.

Most models are wrong. Some are useful. The skill is knowing you’re holding a model instead of absolute truth. Don’t mistake the map for the territory.

🗺️

What useful model could you put to work even if you can’t prove it’s true?

Watch

Listen

“In seed time learn, in harvest teach, in winter enjoy.”

William Blake

My sleep app shows a debt.

Skip an hour of sleep and it adds that hour to the total. A few short nights running and I owe eight hours, then ten. The debt behaves like a loan. Run it up and I get a slow morning and a short fuse.

Bank sleep before you need it and you walk in with a reserve instead of a deficit.

In a sleep lab, one group slept nine hours a night for a week while another slept the usual seven, then everyone dropped to a few hours a night. The nine-hour group held together better. Faster reactions. Fewer of those blank half-seconds where the mind drops what’s in front of it.

A surgeon sleeps long before a run of night shifts for this reason. A new parent who banks sleep in the last quiet week rides out the newborn months better than one who starts empty. The sleep goes in ahead of the demand.

You would feel the same tiredness. Your reactions would just be quicker when the week turned hard.

Sunlight stores and spends the same way.

In summer, your skin makes vitamin D from the sun and stores some of it in body fat. That reserve keeps your levels from crashing the day the light goes short. It holds like a small battery.

The stored form lasts two to three weeks, so a bright July covers a few weeks of winter. A marathon afternoon in the sun banks little more than a short one, since the skin stops making the vitamin once it saturates. People in the far north worked this out and drank cod liver oil when the light thinned.

Animals bank better than I do.

A gray squirrel hoards hardest. Through autumn it buries nuts one at a time across hundreds of holes, a trick called scatter-hoarding: many small caches instead of one pile, so a thief who finds one still misses the rest. Come winter the squirrel digs them up by memory and smell.

Between the raiders and its own forgetting, a squirrel recovers somewhere from a quarter to most of what it hid. The rest gets stolen or sprouts. Those forgotten nuts are how oak forests climb a hillside, planted by a squirrel that meant to eat them.

Your body stores and spends the same way.

A camel carries its reserve on its back. The hump is a mound of fat, a fuel tank for the desert.

On a long crossing it burns that fat and the hump softens and flops to one side, the balance readable from across the dunes. Feed the camel and the hump firms back up, though only to the size it was before.

Fitness banks and drains. Train for months and you build a reserve of endurance. Stop training and it goes: peak aerobic capacity drops about a tenth in the first month off, most of that in the first three weeks.

The body keeps a deposit though. When a trained muscle shrinks from disuse, the nuclei it grew stay put, so the comeback runs faster than the first climb did.

So each reserve comes down to the same three things:

• What each reserve holds
• What caps each reserve
• What each reserve leaks

My own habits are no different.

I bank habits in an app I built for it. On a strong day I do tomorrow’s marketing block early and get the credit. I bank reading too, a session now against a busy week later. A good day pays me twice.

Eat 300 grams of protein in one meal and you build less muscle than eating 30 grams a day for ten days.

Your muscles read a single meal up to about 30 to 40 grams, then the switch that tells them to build shuts off. The rest gets digested and spent elsewhere. Ten daily meals throw that switch ten times. One giant meal throws it once.

One rule I keep is no caffeine after 10am. The rule gives me nothing to bank. It’s worth something only while it’s live today, then again tomorrow.

Over-abstaining on Monday buys no cover for a 4pm espresso on Friday. It takes the same effort each morning and stores no credit for later.

How much you can make up depends on the debt.

• Skip tomorrow’s marketing block and I can make it up by Friday.
• Run up sleep debt and it clears slowly and only part way.
• Years of short nights track with a shorter life.

Deposits fade at their own rate. Some you lay down in a single day and draw on for over a year. Go on a silent retreat or take a heavy dose of psilocybin in a study and you come back changed.

The good feeling fades first. That afterglow lasts somewhere from a day to a month. A retreat’s calm often thins within two weeks of ordinary life.

Part of the change stays for good, though. In one study a single high dose of psilocybin lifted openness, the trait behind curiosity and new ideas. It still read higher fourteen months later. The rest holds only while you keep going back: the change stays while you practice it and drifts when you stop.

Bank where the deposit stays. For the ones that fade, you keep making the deposit as often as it drains.

A habit you can bank lets you aim higher than one you can’t.

Say I count movement in steps.

	Daily target	Why
Without a bank	8,000 steps	low enough that a rough day can’t break the streak
With a bank	15,000 steps	a hard day spends savings instead of snapping the chain

The higher goal only turns safe once a reserve sits behind it.

Without that reserve you live check to check. A sick week, a trip, a day that falls apart: each shock lands at full force. Some of those breaks you don’t come back from, because the streak carried your momentum and your sense of who you are.

Now it’s gone.

Some businesses run on this by design. A fishing town or a mountain lodge earns its whole year in a few good months, banks it, then lives off the reserve through the quiet season. The bank carries the lean months.

A floor low enough to clear on your worst day keeps that day from breaking the streak. A bank lets you set a higher bar, because a bad day spends the reserve instead of breaking the run.

Do today’s rep first. On a day with more in the tank, do tomorrow’s now. A good day can pay you twice.

🏦

What’s one habit you could get a day ahead on today?

Watch

Listen

“People pay for what they do, and still more for what they have allowed themselves to become. And they pay for it very simply: by the lives they lead.”

James Baldwin

I sat down to meditate for one minute a day, fifteen years ago. The sitting now runs up to two and a half hours. Each minute made the next one easier.

A business grew the same way. Over five years ago I started publishing weekly reports on my personal site. Readers kept coming back, so I registered the trends.vc domain.

The company people see today came out of a writing routine, repeated for years.

My outcomes are the results of my daily actions, averaged over time. One day barely counts. A year of days decides the result.

A Post Office official named Anthony Trollope wrote at 5:30 each morning, watch on the table, 250 words each quarter hour, then left for a full day of work. The mornings added up to forty-seven novels. “A small daily task, if it be really daily, will beat the labours of a spasmodic Hercules,” he wrote.

Organizations run on the same math. Toyota’s workers have submitted improvement ideas for over seventy years, around a million a year in recent decades, most from the factory floor and most of them small fixes to the work in front of them. The daily trickle compounded into the biggest carmaker in the world.

Eat processed food each day while believing in health and the food decides the result. Spend each evening scrolling while valuing creativity and you get better at scrolling. The output follows the actions, good or bad.

Your life is the average of your days.

🧱

What daily habits does the life you want require?

Watch

Listen

“The first rule of compounding: Never interrupt it unnecessarily.”

Charlie Munger

Rick was in a hurry.

Warren Buffett, Charlie Munger and a third investor named Rick Guerin were working the same ideas in the same years. The same returns sat in front of all three. Guerin wanted his faster, so he borrowed against his stocks to buy more of them.

That move has a name: buying on margin. The broker lends you money. If your holdings fall far enough, the broker sells you out to get the loan back.

Then the market fell. In 1973 and 1974 stocks lost roughly half their value. Guerin’s margin loans got called. He had to raise cash fast. The one thing left worth selling was his Berkshire Hathaway stock. He sold it back to Buffett at around $40 a share.

Buffett told the story decades later. “Charlie and I always knew we were going to be rich, but we were not in a hurry. And Rick was in a hurry.” Those same shares are worth hundreds of thousands of dollars each today.

Guerin had the returns in reach. He lost his seat at the table.

About twenty-five years on, a fund called Long-Term Capital Management ran a bigger version of the same move. Two of its founders held Nobel prizes. Their trades were sound. To make small edges pay, they borrowed hard: roughly $5 billion of their own capital controlling more than $125 billion in positions.

For four years it printed money. Then in 1998 a handful of markets turned against them at once. In under four months the fund lost $4.6 billion. It took a rescue organized by the Federal Reserve to unwind the fund without dragging banks down too.

The strategy held up. The leverage is what killed it.

My best friend and I were comparing notes on the patterns we’ve each noticed in ourselves lately. He went first. Then he turned the question on me.

I told him mine. I build my days around improvement: I run experiments and I end each night with review questions. What’s my limiting factor? What could I improve right now?

The experiments push toward more and the questions aim the push: try the thing, raise the thing, fix whatever holds the rest back. It’s the same push Guerin felt toward his returns. The questions skip the other half: what am I giving up to get it?

My own portfolio was the example I gave him. A handful of businesses, products in a good spot, marketing as the bottleneck: growth depends on more people finding them. Say the marketing is 30 minutes a day and it’s working. Slow, real growth. Each evening the review returns the same answer: marketing is the limiting factor.

Say I double the marketing to an hour a day. It works. I get there faster.

The extra half hour is a loan, though. Guerin borrowed from a broker. I’m borrowing willpower, effort and attention.

For two or three months the faster pace holds. A year later the marketing is at zero. I overcommitted and the habit collapsed.

The version of me that stayed at 30 minutes is still marketing four years later. Slower than I wanted yet improving the whole time.

The limiting-factor question keeps its place in my evening review. It finds the bottleneck.

When growth is at zero, the bottleneck is a blocker and the move is to solve it.

When growth is already there, just slower than I’d like, the same bottleneck becomes a speed signal: faster is available, at a price.

A habit that’s already paying is compounding. Adding load raises the odds it breaks. So I leave the 30 minutes alone and let time do the work I keep trying to force.

The bottleneck shows you where you can go faster. It hides what faster costs.

⏳

What’s one thing that’s working that you could stop trying to speed up?

Watch

Listen

“Nothing is particularly hard if you divide it into small jobs.”

Henry Ford

One line sat in my experiment queue for weeks: “Set up makemulti.com email address.” What looked like a request for one new inbox was really email setup for two jobs: sending cold outreach from a separate domain and receiving customer email from makemulti.com.

The name undersold the work. I knew it, but hadn’t written the steps down.

I wrote out the work as nine steps:

• Research inbox providers
• Provision the mailbox
• Configure SPF, DKIM and DMARC DNS records
• Buy a separate domain for cold outreach so the primary stays clean
• Provision a second inbox on that domain
• Configure DNS again
• Import into a warmup tool
• Update the sales tool
• Verify deliverability for both cold outreach and customer email

With all nine steps on the page, each had a first action I could take right away. “Research inbox providers” had an obvious next click. “Provision the mailbox” also had an obvious next click.

I could commit to the next step and leave the other eight on the page.

Breaking down the task lowered the resistance by replacing unknown work with a visible list. Choosing one step at a time lowered the resistance again.

In Getting Things Done, David Allen turns multi-step projects into lists of next actions. One task on a list can require several actions.

Write “set up email” and you still have to decide what to do first. Name each step and the first action becomes clear.

Anne Lamott’s ten-year-old brother froze over a school report on birds after putting it off for three months. Her father told him, “Bird by bird, buddy. Just take it bird by bird.”

When you resist a task, check whether it takes several steps. Write out the steps, then choose one you can start.

🪜

Which stalled task could you break into actionable steps?

Watch

Listen

“If you replayed your life and took the shortest path to where you are today, you’d see what an absurd amount of time was spent on the wrong things.”

Julian Shapiro

I ran the replay on my own life and wrote down what I’d tell a younger version of me.

Learn to code by building whole apps. Years earlier. I studied programming from books for years. Books taught me vocabulary: variables and Java objects. Whole-app tutorials taught me to build: follow along, hit errors, search YouTube, make the app run.

Building real things was the only way I learned.

AI writes code now and the advice stands. Knowing how to code shaped how I think and it was my route to financial security.

Believe companies hire off a portfolio. I went from junior to senior in under two years on skills I taught myself. The work is the credential.

Build something real and ship it. Create proof and lead with it.

They can’t deny a portfolio.

Start courage challenges early. I run a daily courage challenge: one act a day that takes nerve.

• The hard conversation I’ve been ducking
• The sales ask
• The pitch a group might turn down
• The call I keep dodging

Anaïs Nin wrote that life shrinks or expands in proportion to one’s courage. Courage is also the root the other things on this list draw from.

Doing these on purpose each day would have gotten me here years sooner.

Treat meditation religiously. I broke my practice many times over fifteen years and each break started the same way: the practice was working, so I poked at it. A shorter sit. A more efficient substitute.

Anxiety crept back a week or two later, slow enough that I blamed something else.

Religious means hands off.

Once a practice proves itself, stop touching it.

⏪

Replay your life. Take the shortest path to where you are today. What do you see?