A week plan written on Sunday evening has a property the founder underestimates: it is going to be wrong. Not entirely, not catastrophically, but wrong in the specific way that Sunday brains are wrong about Wednesday realities. The customer will email. The contractor will reschedule. The investor will surface. The body will catch a cold.

Most founders respond by writing more elaborate week plans, on the theory that better planning will defeat the chaos. The chaos always wins. The better question is how to write a plan that bends rather than breaks.

This page is what such a plan looks like.

The premise

A week plan is not a prediction. It is a commitment structure. The week will not unfold as planned, but the plan determines what gets sacrificed when reality intervenes. A good week plan does not require accurate prediction. It requires explicit priorities, so that the in-the-moment sacrifices are pre-decided rather than reactive.

This is borrowed from the Stoic frame of preparation. Seneca described preparation as the practice of knowing what you would do under conditions you cannot fully control. The week plan is a Stoic instrument: not a script, but a position from which to absorb the unexpected.

For the broader week-architecture this fits inside see how to build a deliberate week.

The first move: name the one thing

Every week has a single most important deliverable. Not three. One.

The one thing is the work that, if it ships this week and nothing else does, makes the week not wasted. For a founder this is almost always the work that compounds: a feature, a customer conversation, the next post, a draft of the next deck.

Naming the one thing is the hardest move in the planning ritual. The brain resists, because naming one thing means de-prioritizing other things, and the founder's brain is built to want everything to be a priority.

The signal that the one thing is correctly named: the rest of the week makes sense in relation to it. If the one thing is "ship the new onboarding", then the meetings that support shipping it are in. The meetings that do not are out, or they happen in the back half of the week. If the one thing is "have ten customer calls", then the calendar is built around scheduling those calls.

If you cannot name the one thing in three minutes, the week is unclear before it starts. Stop and read how to build a deliberate week to see if the missing structure is at the daily or weekly level.

The second move: name the protection

The protection is the thing you will not sacrifice when the week gets ugly. This is usually not work. It is the training block, the dinner with your partner, the phone call to a parent, the sleep window.

Naming the protection is the second hardest move, because the founder's brain wants to keep these flexible "in case the week needs it". The whole point is that they are not flexible. The week needing them is exactly when they protect their function.

The protection is small. One or two items, maximum. A founder who tries to protect twelve items has protected nothing, because the moment the chaos arrives the priority among the twelve is unclear.

The Stoic frame supports this. Epictetus describes the practice of knowing in advance what you will not give up, so that the in-the-moment negotiation is not necessary. The negotiation is the cost. The pre-decision removes it.

The third move: the meetings sort

Meetings break week plans. Not because meetings are bad, but because meetings are scheduled by other people on schedules that do not respect your week.

The sort is binary. Each meeting on the calendar gets one of two labels.

Must happen this week, on the day it is scheduled.

Could move, with low cost.

The "must happen" meetings are the spine. They cannot move. The "could move" meetings are the flexion points. When reality arrives, these are the meetings that flex, not the one thing and not the protection.

Doing this sort on Sunday saves Wednesday from having to do it under pressure. The Wednesday version is worse because the Wednesday brain is tired and reactive.

For the meeting-management principle this depends on see how to handle a shipping week as a solo founder.

The fourth move: the buffer

The buffer is the unallocated time in the week. Most founders have zero buffer in their plans. This is the single biggest reason their plans break.

The realistic buffer for a founder is twenty percent of available work hours. If the work week is forty hours, eight of those are buffer. The buffer is not vacation. It is the unallocated time that absorbs the customer email, the bug that took three hours, the call that ran over, the dependency that was not what you thought.

A week plan without a buffer is a plan that will fail by Wednesday. A week plan with a real buffer is a plan that absorbs the disruption without sacrificing the one thing.

The mistake most founders make is to count the buffer as "available". By Tuesday the buffer is gone because it was counted as workable hours, and the rest of the week is back to chaos.

The discipline is to leave the buffer unscheduled. To say "no" to the meetings that would consume it. To use it for what it is, which is the cost of operating in reality.

The fifth move: the Wednesday checkpoint

The plan written on Sunday gets re-evaluated on Wednesday morning. Not extensively. Five minutes.

Three questions.

Is the one thing still the one thing, given what has happened so far this week.

Has anything in the protection slipped, and if so what is the smallest move to restore it.

What is the new version of the rest of the week, given the actual state on Wednesday.

This is borrowed from the Stoic practice of midday reflection. Seneca described pausing to check whether the day was going in the direction set in the morning. The week-level version of this is the Wednesday checkpoint.

The checkpoint is short on purpose. The brain wants to re-plan the whole week. The whole week does not need re-planning. The plan needs an honest five-minute assessment, after which the rest of the week proceeds in its updated form.

For the daily version of this practice see how to show up for yourself when no one is watching.

What this is not

This is not a productivity system. There is no tool here, no app, no framework with capital letters. The whole point is reduction.

This is also not optimistic planning. The plan does not assume the best case. It assumes the realistic case, which includes disruption, and is built to absorb it.

This is also not pessimistic planning. The plan is ambitious about the one thing. It is just realistic about everything else.

The Stoic register

The Stoics did not write about week plans, but they wrote about the relationship between plan and reality. Marcus Aurelius described the work of the morning as "doing what is set before you, without complaint and without delay". The plan is the "what is set before you". The delay and complaint are what reality forces if the plan was not built to survive contact.

A Stoic week plan is honest about reality. It includes the friction. It includes the buffer. It names what is in your control and what is not. It produces a week that ends with the one thing shipped and the protection intact, even when the days inside it were nothing like the days you imagined on Sunday.

This is what a week plan that survives contact with reality looks like. It is not impressive. It is the structure that does not require impressiveness to function.

What changes

Six weeks of this and the founder has a calibratable week plan. The one thing gets sharper. The buffer gets more accurate. The Wednesday checkpoint gets faster. The weeks where the plan fails entirely become rare, and the weeks where the plan flexes successfully become routine.

The founder who has run this plan for a quarter has data they did not have before: actual records of what got planned, what shifted, what shipped, and what slipped. The pattern across quarters is what the operating layer surfaces. The week is in the calendar. The pattern is in the record.

NothingGiven.

Frequently asked questions

What if the one thing changes mid-week

Note it, finish the original to a stopping point, adjust the rest of the week around the new one thing. The change is data. The system handles it.

What if I do not have a Sunday evening available

Pick any consistent time before the week starts. Friday afternoon works for some founders. Monday morning at 7 a.m. works for others. The day matters less than the consistency.

How long should the plan be

A page, maximum. Most weeks fit on half a page. A longer plan is a plan that will not be re-read on Wednesday, which is when it needs to be re-read.

What if I have multiple urgent things

You still pick one as the one thing. The others are second tier. This is uncomfortable. The discomfort is the point. The brain that has not done the discomfort of picking will spend the week diluting effort across all the urgent things and shipping none of them well.

Is this the same as OKRs

No. OKRs are quarterly. This is weekly. The two can coexist. The quarterly OKR sets the destination. The weekly one thing is what you ship this week toward that destination.

How long until this feels natural

Three weeks of consistency. The first week the planning feels artificial. By the third week the one thing names itself faster than the resistance to naming it.