Plan a week of remote work that actually finishes

Eleven Mondays of "update portfolio"

For a few months I wrote my weekly plan on the left-hand page of a paper notebook every Monday morning. When I finally flicked back through it, one line jumped out: "update portfolio" appeared on eleven Mondays in a row. It never got done, and it never got deleted. Every week it sat under three or four other goals, and every week the client work and the meetings and the admin ate the hours it was supposed to use.

The portfolio was not the problem. The problem was that I wrote outcomes before I had counted anything. Monday-me looked at five empty-looking days and filled them with hope. Friday-me inherited the result.

So this is the order I use now. Count the hours that are already taken, then choose outcomes that fit what is left. Meetings, messages, admin, and the school run are not interruptions to the week. They are the week, and planning as if they will politely step aside is how a goal survives eleven Mondays.

Meetings, messages, admin, and the school run are not interruptions to the week.

The daily version, one Primary and a shutdown, is structure a remote workday. This page is the layer above it: what Friday will be able to verify, and how the days get there.

Count the week before you pick the work

Open Monday to Friday and write four numbers before any goal.

Fixed meetings, including prep. A decision meeting costs the live time plus the minutes to write the question and two options. A status meeting you only attend costs the live time. Do not add a fictional hour of prep to every invite, and do not pretend a two-hour workshop costs two hours when you know you will need twenty minutes afterwards to write up what was agreed.

A message allowance taken from last week, not from a wish. If you have clients or an on-call rotation, that is often thirty to sixty minutes a day, in one or two sweeps. Use last week's real number if it was lighter. Never write zero, because zero means you will answer messages anyway and steal the time from somewhere you did not mark.

Admin you already know, with a duration: invoicing, expenses, a timesheet, a review form. "Sometime Friday" is how admin lands on Saturday morning.

Household gaps and recovery: a pickup, an appointment, a noisy afternoon, a short day after travel. Mark them in the same ink as meetings. They cost the same.

What remains is the deep time you may assign. If meetings take more than half the hours you will sit at the desk, plan one weekly outcome. Why an eight-hour picture of the day collapses is laid out in planning eight hours into eight hours fails; the weekly version is the same arithmetic five times over.

Example. Assumptions: about eight desk hours a day for five days, twelve hours of meetings including prep, one hour a day for messages, three hours of admin, and two hours left unscheduled because last week ran long twice. Forty minus twelve, minus five, minus three, minus two leaves about eighteen hours. That is one weekly outcome and one Secondary, both reviewable slices. Twenty meeting hours would mean one outcome. A result near zero means cut outcomes, not add evenings.

Weekly plan worksheet

Desk hours:
Meetings, including prep:
Message allowance:
Admin:
Household gaps:
Unscheduled recovery:
Deep hours left:
Outcome 1, and how Friday will verify it:
Outcome 2, or blank:
Outcome 3, or blank:
Who must unblock each outcome, by which day:
Buffer you will not fill until Thursday:
Monday first move:

Fill the capacity lines before the outcome lines. If you catch yourself typing an outcome while Meetings is still blank, stop. The sheet is finished when each outcome has a verify test you could apply on Friday afternoon, and Monday's first move is a physical action rather than a mood.

A weekly outcome is evidence. "Staging deploy of the onboarding change, with a demo note in the ticket" can be checked. "Work on onboarding" cannot. "Proposal sent, with two pricing options" can be checked. "Catch up on email" cannot. If you cannot picture the link you would paste on Friday, shrink the outcome until you can. My portfolio line failed this test every single week; "two case studies written and live on the portfolio page" would have passed it, and it would probably have been done in the first fortnight.

Cap the count by the meeting load you wrote down. A heavy live week gets one outcome. A mixed week gets two. A quiet week gets three, and the third is allowed to slip without a crisis. A fourth is a wish, so delete it or move it to a named later week. A list that only grows is not a plan, which is why your task list keeps growing until deletion becomes a weekly step.

Name who must unblock each outcome, and by which day. If two outcomes need the same person on the same day, you have a queue, so sequence them. The ask is a question with a date, not "let's sync."

Why I plan the week and not just the day

Daily plans on their own have a blind spot. Each morning you pick the most pressing thing in front of you, which is sensible, and after five sensible mornings you notice that the important thing nobody was chasing has not moved. Urgent work always looks more pressing at 09:00 than slow work that matters on Friday.

The weekly sheet is what gives the slow work a seat. Outcome 1 gets its deep mornings named on Monday, before Tuesday's surprises arrive, and a new request on Wednesday has to argue with something written down rather than with a vague feeling. That is the whole value. It is not a productivity ritual. It is a short written promise to yourself that you can point at when someone asks for "just a quick thing."

Theme days only where the calendar already agrees

Theme days cut switching. They are defaults that you break on purpose, in writing, on Monday.

When the calendar roughly matches it, a week might look like this. Monday is the plan plus the first deep block on outcome 1. Tuesday morning is deep work on that outcome. Wednesday takes meetings and reviews. Thursday morning is deep again, on outcome 1 or 2. Friday ships what is true, does the admin, and writes Monday's first move.

The trap is choosing a theme the calendar disagrees with. If Wednesday is already a stack of standing calls, there is no point calling Monday "meetings day." Protect Tuesday and Thursday mornings instead, and write the exception into the Monday note: "Wednesday is meetings. Deep work is Tuesday and Thursday before noon." A stated theme is something you can point at when a new invite arrives.

Name the bigger exceptions on Monday too: a release, travel, a school holiday. The exception changes how many outcomes you keep, not just which day they land on.

If you freelance for two or three clients, theme by client. One client gets the early deep mornings, the other gets a later morning and Friday's wrap. I tried running two hard client problems side by side every day for a while, a site migration for one and an app bug for another, and both clients got updates that described activity rather than results. How to lay out a week like that is design a multi-client week without context thrash.

Invoicing goes on the admin block you already counted, usually Friday. A learning goal, if the week can hold one, is a single outcome with evidence, such as "deploy notes are in the ticket." On a heavy week that line stays blank, and that is fine.

Leave buffer you are not allowed to spend on Monday

Buffer is time you can see and do not fill with a fourth outcome on Monday afternoon because the first day went well.

Keep thirty to forty-five minutes on a normal day for overrun or for one reply that genuinely cannot wait. Keep half of Friday afternoon for finishing the main outcome or for an honest wrap. Keep one midweek block for a replan. If that buffer is still empty on Thursday, finish early or pick up something small you already named as If time. Do not promote a new weekly outcome into it.

I learned this the dull way. A client once sent a "small" content change on a Tuesday, I said yes because the week had slack, and the change turned out to touch three templates and a contact form. It took the whole buffer and half of Thursday. Nothing about that request was unreasonable. What was unreasonable was me treating the buffer as spare capacity instead of as insurance for the work I had already promised.

When a request does not fit, answer with the sheet. "This week is locked to the staging deploy, with the demo note by Friday. Your item is first next week." If the request really is an emergency, it replaces an outcome. Say which outcome moved, and to which week. People take that far better than a vague "I'll try to fit it in," which they correctly read as a coin toss.

Run Monday from Friday's note

Monday morning is twenty minutes, then the first deep block.

Open Friday's shutdown and read Done, Open, Blocked, and the first move. If that move is still right, start it before the twenty minutes end. If the weekend added a meeting, change the capacity numbers before you touch the outcomes. A Sunday-evening invite is capacity you did not have on Friday, and shrinking now is cheap compared with shrinking on Thursday.

Confirm one to three outcomes. Put the deep mornings on the calendar before you accept any new invites. A hold called Focus, with an end time, is easier to protect than a private intention, and colleagues generally respect a visible block more than you expect.

Monday's Primary is today's slice of outcome 1, unless the whole outcome honestly fits in a day. Write Primary, Secondary, If time, Waiting on, and First move. Secondary can be a small step on outcome 2. Waiting on names the person from the worksheet. If the twenty minutes turn into fifty, stop planning and start the move.

The loop is simple once it runs: weekly outcomes become daily Primaries, daily shutdowns become Friday's rollup, and Friday's first move becomes Monday's start. Household gaps get marked in the same twenty minutes. If a Thursday appointment removes a deep block, shrink Thursday's Primary now, not on Thursday morning.

Place a week with two clients and a dentist

Example. Assumptions: you are solo, with two clients and no employer standup. Outcome 1 is revised onboarding copy on staging, with a link you can open on Friday. Outcome 2 is a fixed-fee quote sent, with two options in the doc. Outcome 3 is last quarter's expenses filed, and it is the one allowed to slip. The calendar already holds a client call on Tuesday at 10:00, a review on Wednesday at 14:00, and a dentist on Thursday at 11:00. After subtracting those, the deep hours still cover two delivery outcomes if the mornings hold. Messages get a sweep after the morning block, never during it.

Monday's Primary is onboarding sections 1 and 2, then the sweep. The quote waits.

Tuesday: section 3 before the 10:00 call if it fits, questions written, then one forty-minute block for notes. A new section that comes out of the call replaces later polish. It is not a fourth outcome.

Wednesday: paste to staging, then the review. The buffer stays empty unless a fix fits in thirty minutes. A longer fix moves to Thursday, and something else moves with it.

Thursday morning is the dentist. Thursday afternoon is the quote only if staging is done or blocked on the review. Otherwise Thursday stays on outcome 1 and the quote moves to Friday morning. Decide which at midday, and do not try to run both.

Friday morning sends the quote, or finishes outcome 1 if it slipped. Expenses go in the afternoon if they fit. If they do not, they become next Friday's first admin block. Then the weekly shutdown.

The Monday five-line plan uses the same fields as any other day. Example, same assumptions:

Primary: Onboarding sections 1 and 2 drafted in the working doc
Secondary: List the questions for Tuesday's call
If time: Blank
Waiting on: Nobody yet. Review is Wednesday
First move: Open the doc and write the section 1 heading

That is a day. The week is the three outcomes. Mixing the levels is how "send the quote" sneaks onto Monday and the draft never starts.

Place a week that is mostly meetings

Example. Assumptions: you are inside a company, about twelve hours of live meetings are already booked, and you had three candidate outcomes: a checkout decision recorded in the ticket, a roadmap draft someone else can comment on, and performance notes for two people. Twelve meeting hours means three outcomes will lie. Keep two, and say which one waits.

Keep the checkout decision and the roadmap draft. Move the notes to the following week unless a review meeting is already booked for them. If it is, the notes replace the roadmap, and Monday's plan says so in plain words.

If you run the team, try one short recurring sync, a written pre-read, and a two-week trial where one standing meeting becomes a doc. If you do not run it, decline optional invites with no agenda, batch one-to-ones onto a day that is already meetings, and keep two mornings free of new ones. Both versions are in fewer and better remote meetings.

Wednesday replan worksheet

What slipped:
What still finishes this week (one outcome):
What moves to next week:
Friday buffer, used for:
Monday first move, updated:

Use it on Wednesday, or Thursday morning if Wednesday had no hour to think. You need it the moment Monday's outcomes no longer fit the hours left, and most weeks that moment arrives.

One outcome still finishes. The others move, each with a week named. Friday's buffer goes to that outcome or stays free for the shutdown. Rewrite Monday's first move so next week does not inherit a ghost. If "still finishes this week" has two items, you have copied Monday's list rather than replanned. Cut an outcome before you cut sleep.

A slipped outcome that you will not schedule next week gets crossed off rather than carried. This is the step my portfolio line never got. Carrying it was easier than admitting it was not a priority, and so it sat there for eleven weeks making every plan look slightly fuller than it was.

Release week: one ship outcome, Secondary frozen. Holiday: handoffs before you leave. Travel: one outcome plus admin. If you delegate, each outcome has an owner and your job is the Friday verify. Quarterly goals stay off this sheet until one of them can be verified on a Friday.

Friday status update

Done:
Open:
Blocked:
First move:

Set a timer for fifteen minutes and answer a few questions before writing the four lines. Which outcomes have a link someone else can open? What carried over, and is it still worth carrying? Which theme day lied, the calendar or the outcome count? What is Monday's deep-block Primary, in one line? What gets deleted?

• Which outcomes have a link someone else can open
• What carried over, and is it still worth carrying
• Which theme day lied, the calendar or the outcome count
• What Monday's deep-block Primary is, in one line
• What gets deleted

Done is evidence. Open includes a size. Blocked names who, the ask, and the day. First move is Monday's physical action, so "continue the roadmap" gets rewritten to the file and the heading. The daily version, including what a client should and should not see, is the end-of-day status update.

Example. Assumptions: the onboarding copy is on staging, the quote was sent, expenses slipped, nobody is blocking Monday, and client comments are expected then.

Done: Onboarding copy on staging (link). Fixed-fee quote sent (link).
Open: Expenses, about an hour. First admin block next Friday.
Blocked: None. Client comments on the staging doc are expected Monday, not a blocker today.
First move: Open the staging doc comments and turn them into a list of edits.

Put the note where Monday will find it. On a crunch week, spend ten minutes instead of fifteen and delete harder. A system that only works in a quiet week fails exactly when you need it, which is the pattern in why productivity systems fail after the first week.

Friday afternoon is not for starting things

The last habit is the one I resisted longest. When an outcome finishes early on a Friday, the instinct is to start next week's work right away and get ahead. I used to do this, and it almost always backfired: the new thing was half-started at 17:00, it had no plan behind it, and on Monday I opened it with no memory of why I had made the choices I made.

Now an early Friday finish goes to admin, to the shutdown note, or to stopping early. Next week gets planned on Monday with fresh numbers, not on Friday with leftover momentum. A week that ends with two verified outcomes and a clean note is a good week, and it does not need a bonus half-task bolted onto the end to prove it.