When the day goes wrong by ten o'clock

Updated 2026-09-12

The short answer

Re-plan once, against the hours that are actually left.

The common failure is not the emergency. It is carrying on with a plan built for a day that no longer exists, falling further behind it all afternoon, and finishing with the sense of having failed at something that stopped being possible at ten o'clock.

Three minutes: what is left, what still fits, what is now not happening. Then tell whoever is affected by the third item.

Adding up the estimates is the test: if the total exceeds the hours, no ordering rule fixes it HOURS ACTUALLY AVAILABLE what is left of the day WHAT YOU PLANNED, ADDED UP the shaded overhang is a capacity problem, not a prioritisation one. Putting the right things first still leaves the same number of hours.
Adding up the estimates is the test: if the total exceeds the hours, no ordering rule fixes it

Why the old plan has to go

It was built against a number of hours you no longer have. Every item on it is still sized for the morning version of the day, so working through it produces a growing gap between plan and reality — and that gap is what makes the afternoon feel like failure rather than like a shorter day.

The arithmetic changed, so the plan must. Four hours of intentions do not fit in two hours, regardless of how they are ordered. Recognising that at ten is worth more than discovering it at five.

The three-minute reset

  1. Count the hours actually remaining. Honestly, after the thing that went wrong is dealt with, and after the meetings that are still happening.
  2. Pick the one thing that most needs to be true by this evening. One. Not three — the day already lost its margin, and a reset plan that is itself overcommitted fails the same way the first one did.
  3. Name what is now not happening, specifically, and say so to whoever planned around it. This is the step people skip, and it is the one that costs most when skipped, because by tomorrow it is a surprise rather than information.
  4. Everything else goes back to the list, unchosen and unmarked. Not failed — not chosen.

Distinguish the two cases

Something changed. A genuine emergency, a shifted deadline, a dependency that resolved or collapsed. Re-planning is correct and the plan was not wrong; the world was different.

Something arrived. Loud new work that feels urgent because it is new. That is not a reason to abandon a plan — it is a reason to put the item on the capture list and consider it at the next fixed point. Most of it turns out not to be urgent once it has waited ninety minutes.

The distinction matters because most lunchtime collapses are arrivals wearing the clothes of changes. Ask which happened before you re-plan; if it was an arrival, the answer is to carry on.

One reset, not continuous re-planning

Re-plan once. Then hold it.

A day that gets re-planned three times has no plan at all — each reset costs a switch, and the ordering stops carrying any information about what matters. The value of the reset is that it produces something stable to work against for the remaining hours, and that value disappears entirely if it happens again at two o'clock.

If the day genuinely breaks a second time, that is two events, and it is a day to salvage one thing from rather than to keep reorganising.

The part that is not scheduling

Worth saying plainly, because most advice on this skips it.

A day that goes wrong early is not evidence about you. It is one day in which something unexpected happened, which is the normal condition of working. The most common expensive response is not bad prioritisation — it is spending the afternoon in a low-grade attempt to catch up on a plan that is no longer achievable, which produces less than a calm two hours on one thing would have.

An unfinished task is unfinished. The calendar moving on does not convert it into a failure, and a tool that marks it that way has decided something it has no standing to decide.

Where to go next

Using this in Decide One

Decide One's day is small enough to rebuild in a minute. Overrunning adds the time and says nothing about it, and a line left unfinished carries no mark — which is what makes a reset feel like re-planning rather than like admitting something.

A line you did not finish is unfinished. It is not marked against you, and a day you did not open is not counted.

Open Decide One — free to use, with no account. Nothing you write leaves your device.

Decide One is a priority instrument. Bring the day into view, choose a method, give the first thing real time.

Open it — free, nothing to sign up for