How I Run

This is my operating system. Not aspirational. It describes what I actually do and what I come back to when I drift.

The goal is not to work hard. It is to build a life where the goal is a foregone conclusion. The habits, the structure, the protected time — none of it is the point. They are the conditions that make the outcome inevitable. Someone following me for a day should not need me to explain what I am working toward. They should be able to see it.

Principles

The cadence below is how I spend the hours. These are the rules underneath it, the ones I fall back on when the schedule breaks.

Heads down. During the sprint there are no calls and no messages. The hardest thing gets my whole mind or it does not get done, and the world can wait ninety minutes.

Anchor the morning, not the night. The day is built from its start. I fix the wake and let the bedtime fall out of it. Negotiate bedtime and everything downstream drifts.

Capture fast, decide later. The thought leaves my head in seconds and gets sorted after. Any friction between a thought and a safe place to put it is a leak.

One fate per thing. Every message gets a single outcome before I move on. Half-handled is worse than untouched, because I only have to read it again.

Never hold a monkey. Every problem has a next move, and that move sits on someone’s back. I leave no conversation without a named owner and a named next action.

Activity is not outcome. When someone is fluent about what they did and vague about what changed, the outcome is in trouble. I follow the vague part and leave the fluent one alone.

Feedback same day. I name the behavior, the impact, and what I want more or less of, while the moment is still warm. Delay strips it of context and of use.

Do the work only I can do. The test of a week is not how much moved, but how much of what moved actually needed me. The rest is a candidate for delegation.

Favor what compounds. Between two uses of the same hour, I take the one whose returns build on themselves. A one-off task is spent the moment it is done.

Plug the leaks. If a thing does not return something proportional to what it costs, it leaves the system. Small drains do not hurt once, but stacked they pull me below the line I am capable of.

Stay upwind. Given a choice, I favor what keeps more doors open. A commitment that narrows what I can do next costs more than it looks.

Hold the plan loosely. I walk in with the day fully planned and fully willing to have it rearranged. Certainty about how the day will go is the surest way to be wrong about it.

Two-way doors, walk through. If I can walk back through the door, I do not ask permission and I am not needed in the room. Only the irreversible comes to me first.

Light before the ledger. I cannot run the audit with the mind that is under audit. The correction has to come from outside me first, before my own write-up of the day is worth trusting.

Daily

Fixed wake, same time daily. The day has one anchor, and it is the morning, not the night. I wake at the same time seven days a week and do not let it slide to absorb a late night. Everything downstream holds only if the start is fixed: the habits, the sprint, the batched passes. So I anchor the wake and let the bedtime fall out of it, instead of negotiating bedtime late at night when I always feel fine.

Habits first. Workout, quiet time, and reading. Every day, before work. Non-negotiable. Not productivity hacks. They are what I am built on, and what I come back to when everything else gets loud.

Fast days: Tuesday and Wednesday. I fast from 5am to 6pm. The day bends to protect both the fast and the sleep anchor. A small meal before 5am, then back to sleep, so the wake time never moves. Training stays light on these two mornings, since I do not refuel until evening. I break the fast at 6pm on purpose, not whenever the meetings happen to end, so the first meal never lands against bedtime.

Morning: walk the day before I walk it. Quiet time is not only reading. I lay the day’s parts in front of God and pray over each one. Where it could go wrong, where I will need patience, where I am likely to lean on my own strength instead of His. I plan the day, but I hold it loosely. The plan is mine to make, the establishing is His. I do not just ask Him to show me what is coming. I ask for the strength to go through it.

Deep work: one 90-minute sprint. The brain runs on ultradian rhythms, roughly 90 minutes of high focus then a real break. I do the hardest, most important thing first. No calls, no messages during the sprint. The break after is not optional. See Manage Your Energy, Not Your Time.

Lately I have started measuring the rhythm instead of trusting my read on it. Heart rate, stress, and Body Battery logged through the day, surfaced as a heat-map over my planner. The peak I drop the sprint into is the one the data shows, not the one I assume.

Batch messages, twice. One pass through email and Slack after the sprint, one before closing. Not reactive. Scheduled. Each pass clears the channel rather than skims it: every message gets one fate before I move on. Replied, if it takes two minutes. Captured into my task list, if it needs more. Handed to its owner, if it is theirs. Archived, if it needs nothing. The inbox is not a to-do list. The moment an obligation lives in an unread message instead of my task list, it is already buried under the next ten. A thread I leave open is a monkey I have absorbed. See Marc Andreessen on productivity.

Plug the leaks. Know what is draining capacity and remove it. A late evening meeting that could have waited until morning. Staying out longer than necessary. A commitment accepted because it was easier than saying no. None of these cost much once. Together they compound into a version of yourself operating below what you are capable of. The standard is: if it is not returning something proportional to what it costs, it does not belong in the system. See Plug the Leaks.

Calls and meetings. Mid-day and afternoon. 1:1s, product calls, external conversations. Some Systeric calls run into the evening given time zones. I try to be heads-off by 8:30pm.

End of day: clear the deck, place the monkeys. Every open item has a named owner and a named next action before I close the laptop. I do not end the day carrying monkeys I should have handed off. Read Who’s Got the Monkey.

Shutdown on a trigger. The morning has an anchor and the night needs one too, or it drifts until I crash. An alarm starts the wind-down before lights-out: work stops, conversations end, the phone goes face down. Then the deck-clearing above and the examen below, then sleep at a set time. The last stretch of the day is decompression, never the most activating thing in it. A live conversation or an open problem does not get the slot right before bed, because it is the one thing certain to keep me awake in it.

Before sleep: the examen, ten minutes. A daily error log, not a journal. What did I predict yesterday that did not happen? Where did I waste effort or spring a leak? Where did an assumption go wrong? I open it with one line of prayer before I write anything: search me, show me what I am not seeing. Light before the ledger. I cannot run the search alone because the thing doing the searching is the thing under review. The log is where the flesh’s autopilot gets caught on tape, the habits that keep operating out of reflex even after the heart has changed. One focus point at a time, patience right now, worked until it is habit, then the next. When the same fault shows up for the tenth time, the move is to ask, not to grit harder. The review ends in a return to God, not just a tighter plan for tomorrow. That resolve becomes the next morning’s walkthrough. See Search Me.

Weekly

Monday: one priority question. What must move this week that only I can move?

Friday: week review. One paragraph. What moved, what did not, and why. One question: did I spend most of this week in work only I can do? If not, what shifted and why.

Thinking time, when needed. When direction feels fuzzy, when a pattern across 1:1s will not leave me alone, when a quarter is approaching and I am not clear yet, I block time to be alone and quiet. I bring what I have been collecting and let it surface and connect. Some of it becomes decisions. Some becomes questions I want to bring to specific people. Some dissolves. This is where direction comes from before it becomes a plan. It is also where I pray and listen, not just reason. What I am protecting when I do this is not just reflection time. It is curiosity — the condition that makes original thinking possible. The moment curiosity gets crowded out by the operational load, direction becomes reactive.

1:1s

Capture is fast; the 1:1 is thoughtful. These are opposite tempos and I do not cross them. Capture is a reflex: get the thought out of my head in seconds, before it is lost or breaks my focus, and sort it later. Any friction there is a leak. The 1:1 is the reverse, the one place I deliberately slow down, because the thing worth having sits under the report and only surfaces when I am present and unhurried. Rushing a 1:1 turns a person into a status queue, and the status was already in the doc. The failure most people run is inverted: fast with people, slow with capture. I want it the other way.

Biweekly, with every direct report across engineering, product, and marketing. They fill out the template before we meet. I read it in advance. The meeting is for solving problems, not reading updates.

A good biweekly ends early because everything got resolved. They came in with blockers; they leave without them, or with a named date. I came in not knowing where the quarter actually stands; I leave knowing. The takeaway for them: at least one decision made, one blocker cleared, one piece of feedback they can act on today. The takeaway for me: a real read on what is happening, not what the doc says. Three things I watch for every time: are their goals on track, what is blocking them that I can remove, and are they giving and receiving honest feedback.

Probe for the real state, not the report. The template tells me what they did. The meeting is for what is actually true, which sits a layer under that. So I do not ask how it is going. I ask what is hard about it, what would make it slip, and what they would do if I were not in the room. When someone is fluent about activity and vague about the outcome, the outcome is in trouble, so I follow the vague part and leave the fluent one alone. I am not finished until I can say the real state back in one sentence and they tell me yes, that is it.

Diagnose the gap before I reach for a fix. Anyone stuck is in one of four gaps, and each takes a different fix. A clarity gap: they do not know what good looks like, so the fix is a sharper target. A skill gap: they know the target but cannot hit it yet, so I teach, pair, or shrink the step. A blocker: they know and they can, but something outside them is in the way, and clearing it is often the only thing I can uniquely do. A will gap: they know, they can, nothing blocks them, and they are still not driving, which is a stakes conversation, not a help conversation. The whole skill is not mistaking one for another. Coaching a blocker is useless, and clearing what is really a skill gap just teaches helplessness.

What I press on changes with the seat. The diagnosis is the same for everyone. What I probe, and what good looks like, is not. With an engineer I press on the hardest technical decision and whether they are stuck on ambiguity or on execution, and good is a shipped capability at a quality bar. With a PM I press on whether they can name the user problem, the one metric that matters, and the tradeoff they are making right now, and good is an outcome moved, not a document produced. With marketing I press on the audience, the message, and the number they are moving, and good is a measurable result tied to a channel. For all three the push is the same, from activity to outcome, and then which gap stands in the way.

Targets that survive the walk back to their desk. A target is a deliverable, a number, a date, and a reason it matters. Drop any one of the four and it is a wish, not a target. The test is whether they can say it back without looking at notes; if they cannot, it is not set yet. For someone who can find the path I set the outcome and leave the method to them, because handing a capable person the method is a leak. For someone still building the skill I set the next step instead, because handing them only the outcome is abandonment.

Quarterly, I come in with direction. Before the conversation, I set direction for each person. Two sentences: where I need them to be by end of quarter, and why it matters for the team. Not a task list. A directional statement: what the team needs them to own, or what capability they need to unlock. I write this before the meeting, not during it.

They come with their proposed goals, filled out in advance. I come with direction. The conversation is the alignment between the two. For each of their goals: does it connect to the direction? If yes, sharpen it: deliverable, number, timeline. If no, name why and redirect. If the direction requires something not in their list, name it together and add it.

Three goals. Each with a deliverable, a number, and a reason it matters. They leave knowing exactly what they are building and why. If they cannot say it without looking at notes, the goals are not right yet.

Monthly

Code walk with engineers. How I stay close to engineering quality without creating a bottleneck.

External thinking call. One or two a quarter. A short list of advisors and peers whose thinking sharpens mine. I come with a question, not an agenda.

Quarterly

Energy and zone of genius audit. What gave me energy in the last 90 days, and what drained it? The answer shapes the next quarter’s calendar. The question underneath: am I spending time on the work only I can do, or filling my calendar with things someone else could handle?

Direction. Before each quarter begins, I need to be able to state where I want the team to be by the end of it in two sentences. If I cannot, I have not thought about it enough. When choosing what to commit to, I favor options that keep more doors open over ones that foreclose. A commitment that narrows what I can do next carries a cost that is easy to underestimate. Stay upwind.

People

Most decisions do not need me. I trust the team to think critically and handle things. Most decisions are reversible. The team makes them without me. The two-way door principle: if you can walk back through the door, walk through it. Do not ask for permission. Irreversible decisions come to me before they are made. These include architectural choices that are hard to unwind, new vendors with significant lock-in, material changes to core user flows, and anything with lasting financial or product consequences.

For product and tech calls: I join when the decision is irreversible or when I can see something across contexts that the team in the room cannot. If neither is true, I stay out.

Feedback: same day. When I notice something worth naming, I say it the same day. Delayed feedback loses context, and the person cannot act on it at the right moment. I name the specific behavior, the impact, and what I want more or less of. No hedging.

For people who need more support: Some people are overwhelmed, not operating at the level I expect, or have not yet internalized the direction. For those people, I increase the cadence: daily 15-minute check-ins for a period, until they are operating with more confidence. This is not performance management. It is mentoring. The goal is to close the gap, not document it.

Where each person stands sets the cadence. The biweekly is the baseline, but the read on where someone sits against their level is what decides whether to hold it, raise it, or change what it is for. That read, and how it picks the instrument for someone under, meeting, or over their bar, is its own framework: Performance Management.

What I Do Not Hold

Monkeys. Every problem has a next move. That move lives on someone’s back. When I leave a conversation without a named owner and a named next action, I have absorbed a monkey or lost one. Every conversation ends with someone owning what happens next. Not “let’s follow up.” Who follows up, on what, by when. Read Who’s Got the Monkey.

Decisions that are not mine to make. If a decision can be made without me, it should be. My job is to build the conditions where most decisions happen below me. If the team needs me for everything, I have not built context. I have built dependency.

What I Read

Books I return to:

• The Bible
• Scaling People (Claire Hughes Johnson)
• High Growth Handbook (Elad Gil)
• CEO Excellence (Carolyn Dewar et al.)
• Smartcuts (Shane Snow)
• Problem Solving 101 (Ken Watanabe)
• The Mochary Method (Matt Mochary)
• Rituals of Great Teams (Shishir Mehrotra)
• Rituals of Hypergrowth (Shishir Mehrotra)

Articles worth re-reading once a year:

• Who’s Got the Monkey — Oncken & Wass. The delegation trap, clearly named.
• Managing Oneself — Peter Drucker. Know your strengths, how you work, what you value.
• Manage Your Energy, Not Your Time — Schwartz & McCarthy. Ultradian rhythms, energy renewal, why sprints beat marathons.
• Marc Andreessen on productivity — Scheduling, reading, output.
• pmarchive on personal productivity — The async and batching logic.
• How to Do Great Work — Paul Graham. Curiosity as the core, staying upwind, and protecting the conditions for original thinking.

The Test

At the end of each week: did I work in my zone of genius? Did I protect the things worth protecting? Did the people I invest in leave better equipped than before? Is my direction for the quarter sharp enough to say in two sentences?

If yes, it was a good week. The output follows.