The first seven receipts: a one-week experiment

23 August 2026 · 6 min read

The most useful time tracking experiment takes one week, costs nothing, and has exactly one rule: write down what you think your days look like before you see the first receipt. Then let a Mac record seven days without you touching a timer, and compare. The gap between the guess and the receipts is the result, and it is usually large enough to change how you plan the following week.

Why seven days, and why guess first

A week is the shortest period that contains every kind of day you have: the meeting-heavy one, the head-down one, the one that falls apart. Shorter than that and you are measuring a single day’s weather. Longer and people stop before the end.

Guessing first is the part most people skip, and it is the whole experiment. If you do not write the guess down, your memory will quietly revise it to match the receipts and you will conclude “I knew that.” You did not. Why your estimate of your own workday is wrong by hours explains why the revision happens; the written guess is how you catch it.

Set it up

Take a sheet of paper (paper, so you cannot edit it later) and write five lines: - Hours I work on a typical day: ___ - The app I spend the most time in: ___ - The app I spend the second most time in: ___ - Hours per day in email and chat combined: ___ - The weekday that is my most productive: ___
Pick a closing time. The hour you intend to stop. Not the hour you usually stop; the hour you intend to. 6 p.m. is common.
Install Punchcard and set that closing time. It takes about a minute: there is no setup wizard, no projects to create, no account. It asks for no macOS permissions at all, so there are no dialogs to click through. How to set up Punchcard in ninety seconds has the exact steps if you want them.
Put the sheet of paper face down somewhere you will not see it for a week.
Work normally. This is the hard part.

Punchcard’s first seven receipts are free, which is not a coincidence. Seven receipts is one working week plus the weekend, which is the length of this experiment. Tracking itself never expires; only the receipts after the seventh need a license.

What to do each evening

At your closing time, a receipt prints. Paper feeds out in the menu bar over a couple of seconds, each app on its own line with its time, a total, and a stamp. Read it once, top to bottom, and do nothing else. Do not change your behavior because of it. Do not look at the sheet. Tear it off if you like (drag down, it rips like paper) and go do something else.

If you want a record, export the PNG or copy the plain text into a note, one per day. You will want all seven in front of you on the last day.

Two things to resist:

Adjusting mid-week. If Wednesday’s receipt shows three hours in chat and you decide to fix that on Thursday, you have contaminated the measurement. Fix it next week. This week you are only watching.
Explaining it away. “Tuesday was unusual.” Every day is unusual. Write the explanation on the back of Tuesday’s receipt if you must, and keep going.

Day seven: turn the paper over

On the seventh day, lay out all seven receipts and turn the sheet of paper face up. Go line by line.

Hours per day. Add up the seven day totals and divide by the number of days you actually worked. Compare with your guess. The gap is usually an hour or more, and the direction is the interesting part. If the receipts are longer than the guess, you are working more than you think, most likely in the evenings. If they are shorter, your days have more slack in them than you believed. Both are worth knowing.
Top app. Was it what you wrote down? For many people the honest top line is a browser or a mail client, and the “real work” app is second or third. That is not a moral failing. It is information about what the job currently is.
Email and chat. Add the lines for those apps across the week and divide by days. This is the number people are most wrong about, typically by guessing low.
Most productive weekday. Look at which day had the biggest number next to the app you consider your actual work. If it matches your guess, you know yourself well. If it does not, move your hardest task to the day the receipts picked. How to find your most productive hours takes this further.
Write one sentence at the bottom of the sheet: the single thing that surprised you most. Keep the sheet.

What the receipts will not settle

Punchcard sees app names only. It never reads window titles, document names, URLs, or anything on screen. So the receipts will tell you that the browser had two hours, and they will not tell you whether that was documentation or drift. If your work lives in a browser, the experiment still works, but you are measuring “browser” as one block. A practical fix is to use one browser for work and a different one for everything else, so they appear as two lines.

There are no projects or clients on a receipt. If your question is “how much of the week was for client A,” this experiment does not answer it, and a timer-based tracker with tags is the right tool for that question.

And the experiment measures one week. A week with a conference, a holiday, or a launch in it will not represent you. If that was this week, run it again next week.

The week after

The experiment ends with a decision, not a feeling. Pick one line on one receipt and change it next week. Not everything: one thing. Move a meeting, close the chat app until 11, start the day in the editor instead of in mail. Then let the next seven receipts print and see whether that line moved. How to tell if a productivity change actually worked is the follow-up for that second week.

Questions

Does this work if I use a timer-based tracker instead? It can, but the experiment depends on a record you did not have to remember to create. If you forget to start the timer on Wednesday, you have six days and a hole. Automatic recording removes that failure mode, which is why this post assumes it.

What if I work across two Macs? Each Mac records its own days and prints its own receipts. For the experiment, add the two totals together each evening. A Punchcard license covers two Macs, though for the free week you do not need one.

I got to day seven and the receipts look fine. Did I waste a week? No. “My days look the way I think they look” is a real result, and a rare one. You now have a baseline, which is what makes the next change measurable.

Should I tell my team I am doing this? There is nothing to tell. Punchcard records on your Mac only, sends nothing anywhere (it contains no networking code), and shows the receipt to you. If you later want to post a receipt, that is your choice.