The Sunday roll: a weekly review you will not skip
· 6 min read
Most weekly reviews die because they ask you to assemble the week from memory, calendars and task lists before you can review anything. A review you will not skip is one where the week arrives already assembled: a single roll, printed on Sunday, that you read in five minutes and answer three questions about. Punchcard prints one automatically. A notebook can do the same job, if you wrote in it every day.
Why weekly reviews get skipped
Ask anyone who has tried a weekly review why they stopped and the answer is nearly always the same: it took too long, and the first twenty minutes were spent working out what happened.
That is the structural flaw. A weekly review has two halves, gathering and thinking, and gathering is the tedious half. Scroll back through the calendar. Open the task manager. Try to remember Tuesday. Check the commit log or the sent folder. By the time the week is in front of you, the energy for thinking about it has gone, and the review becomes a list of things you did not finish, which is not a review, it is a reproach.
If the gathering half disappears, the review becomes five minutes of reading and three questions. That is a ritual you can do on a Sunday with coffee, and you will not skip it because there is nothing to dread.
What the Sunday roll is
A roll is a receipt for a longer period. Where a daily receipt itemizes one day by app, a week roll covers the seven days: what the week’s hours went to, and the total. It is the same format, in the same paper style, printed at the end of the week the way a shop prints its weekly takings.
Punchcard prints one every Sunday. During the week it sits in the menu bar and notices which app is in front, by app name only; it never sees window titles, document names, URLs or keystrokes, and it asks for no macOS permissions at all. Each day at your closing time it prints a daily receipt, and on Sunday the week roll feeds out of the same slot, line by line, with the dot-matrix whirr and the stamp landing last. Tear it off. Export it as a PNG or copy the text if you want it somewhere else.
What makes the roll work as a review is that you did nothing to produce it. There was no gathering. Sunday arrived and the week was on the desk.
It is worth saying what the roll is not. It is organized by app, not by project or client, so it will not tell you how many hours went to a particular piece of work. It does not grade the week or compare it with a target. It has no team view. For a review that needs per-project hours, it is the wrong instrument; for “what did this week actually consist of,” it is the right one. What a week of workdays looks like as receipts shows a typical one.
The five-minute read
Here is how to read a week roll (or seven daily receipts laid in a row) without it turning into a project.
- Read the total first. One number for the week. Say it out loud. Is it more or less than you would have guessed before looking? Most people guess low, by several hours; why your estimate of your own workday is wrong by hours explains why.
- Find the top line. The app with the most time. Is it the one where your real work happens? If the top line is the chat client or the browser, that is the single most useful fact the roll will give you.
- Find the longest day. If you have the daily receipts, lay them out and look for the tallest total. Ask what that day had in it that the others did not.
- Find the surprise. There is nearly always one line you did not expect: an app you forgot you used, or one that took twice what you thought. Circle it, literally if the roll is printed.
- Stop reading. Five minutes. The roll will still be there next week, and the review is about the three questions, not about the data.
The three questions
After the read, answer these, in writing, one line each. Paper is best, because paper cannot be edited later and that is part of the honesty.
What took more time than it deserved? Not what took the most time; what took more than it earned. The chat client may deserve its two hours. The browser may not deserve its seven.
Which day do I want more of? Look at the daily totals and the shape of each day. One of them was the good one. What made it good, and is any of that repeatable?
What is the one change for next week? One. Not a plan, not a system, not a new tool. “Close the chat client before lunch.” “Start the big task on Monday, not Thursday.” Write it on the roll itself, next to the stamp, so next Sunday you can check it.
That is the whole review. Three lines and a circled surprise. Over a month of Sundays those lines become a record of what you tried and whether it worked, which is something almost nobody has about their own working life. How to tell if a productivity change actually worked is about reading that record.
Doing it without Punchcard
None of this requires an app. It requires a week that has been recorded as it happened, rather than reconstructed on Sunday.
A paper log works: at the closing time each day, write the date, a total, and three or four lines for where the hours went. Keep the pages together. On Sunday, lay out the seven pages and follow the five-minute read above. The weakness is that the daily writing depends on remembering to do it, and on remembering the day accurately when you do, which is hard at 6:00 p.m. after a long one. The strength is that you chose the lines, so they can be projects rather than apps.
Screen Time on the Mac can give you the app totals for the week, though it counts everything the Mac was used for, including the evenings, and it is a dashboard rather than a record: nothing arrives on Sunday unless you go and look. If you want to try that route, how to use Screen Time on a Mac for work covers it honestly.
The common factor is that the gathering has to be automatic, or it has to be daily. The one thing that does not work is gathering on Sunday.
Questions
What if I missed a few days during the week? The roll still prints with whatever was recorded. A week with gaps is still a week, and the gaps are themselves information: they tend to be the days you were away from the Mac, which is usually fine. Do not reconstruct the missing days; review what is there.
Should I keep the rolls? Yes, at least for a month. Punchcard also prints a month roll, and a stack of four week rolls next to the month roll is the best five-minute picture of a working month you can get. After that, keep the ones with a circled surprise and let the rest go.
Does the review have to be on Sunday? The roll prints on Sunday, but the reading can be Monday morning before the first message arrives. What matters is that it is once a week, at the same point in the week, with nothing to gather first.