Plan a week that actually happens

A to-do list tells you what to do, a calendar tells you when, and only the calendar survives a busy week

You are at the Calendar step of the Empty-Head Loop. You have a clean list of next actions, but a list is a wish until it has a time. A weekly planning system turns the processed work into time blocks on a real calendar, and the rule that makes it hold is brutal honesty about capacity. The shift here is that a scheduling agent places the blocks and reshuffles them when Monday breaks the plan.

By the close you will have a time-blocked week with protected deep-work blocks that survives contact with reality, and an agent keeping it current as the week moves.

Time-blocking turns intentions into appointments

Time-blocking is assigning each piece of real work a specific block on the calendar, so the calendar, not the list, drives your day. A list lets you lie to yourself about how much fits in a day; a calendar forces the truth, because a day has a fixed number of hours and a block either fits or it does not.

Block by context and energy, not just by task. Cluster shallow admin into one batched block, give deep work the hours your brain is sharpest, and put fixed client commitments down first as immovable anchors. The processed action list from the last chapter is your raw material: each action, with its context tag and estimate, becomes a block.

The 60% rule: plan to the capacity you actually have

The single reason most weekly plans collapse is that people plan to 100% of their available hours, leaving no slack for the inevitable. A client emergency, a call that overruns, a task that took twice its estimate, and the whole sequence falls like dominoes.

Plan to 60% of your working hours. If you have a forty-hour week, schedule about twenty-four hours of committed blocks and leave the rest as buffer. This is not lazy, it is the only number that survives a real week, because the unplanned forty percent is where the overruns and the genuinely urgent live. A plan built to 60% bends and recovers; a plan built to 100% shatters on the first surprise.

Let the scheduling agent place the week

Manually fitting blocks around fixed commitments, then reshuffling every time something slips, is exactly the tedium to delegate. Motion is built for this: it reads your task list with durations and deadlines, reads your fixed calendar events, and auto-places every task into the open time around them. When a meeting runs long or a task slips, it reschedules the rest automatically rather than leaving you to redraw the day by hand.

The exact steps:

1. Feed the agent your processed actions with a duration estimate and, where it exists, a deadline. The estimates from your triage agent feed straight in.
2. Mark your fixed anchors as hard calendar events the agent must schedule around: client calls, the school run, a standing commitment.
3. Declare your deep-work window as protected hours the agent fills only with deep tasks, never admin.
4. Set capacity to roughly 60% by capping the daily committed hours, so the agent leaves buffer rather than cramming.
5. Let it auto-schedule, then sanity-check once. The agent proposes the week; you scan it for anything that feels wrong and adjust, rather than building it from scratch.

Motion auto-scheduling a week: a worked example

Consider a solo copywriter running a one-person studio with three retainer clients. The challenge was plan-versus-reality drift: a hand-built weekly plan looked perfect on Monday morning and was meaningless by Tuesday lunchtime, because one overrunning client call knocked every later block out of place and the copywriter never rebuilt it, so the rest of the week ran on reactivity.

The approach was to move scheduling to Motion. Tasks went in with durations, the three retainer calls and the gym were marked as fixed anchors, mornings were declared the protected deep-writing window, and daily capacity was capped to leave buffer.

The measurable result across a month: before, the copywriter completed roughly 55% of Monday's planned blocks by Friday; after, completion rose to about 85%, because every slip triggered an automatic reshuffle of the remaining work instead of a collapse. Deep-writing mornings, previously the first thing sacrificed, were protected four days in five.

INTERVIEW EWOUD: Tell me about a week where your plan held all the way to Friday versus one where it collapsed by Tuesday. What was different about the week that held?

Pitfalls at the planning stage

Planning to 100%. The most common and most fatal error. Build to 60% or the plan dies on the first surprise.

Leaving deep work unblocked. If deep work is not a protected block, the shallow urgent will eat the hours. Anchor it first.

Over-trusting the auto-schedule. The agent places blocks well but cannot feel your energy. Sanity-check its proposal once before committing.

Re-planning from scratch when it slips. The point of the agent is that you never rebuild; you let it reshuffle and carry on.

With a week blocked to a realistic capacity and an agent keeping it current, you have a plan worth defending. But a plan only holds if the blocks survive the noise of a connected day. Next you cover those blocks, with a noise-filter agent guarding your attention.