Where a founder's week actually goes, and how to find out

24 September 2026 · 6 min read

Most founders describe their week by the work they believe matters most: building the product, selling it, raising money. A week recorded on the Mac usually tells a different story, with mail, Calendar, a chat app and a spreadsheet near the top and the product itself further down than expected. You can find your own split in one ordinary week without building any system, and the useful part is not the number but deciding which of the gaps you actually want to close.

The job you describe and the job you do

A founder’s job changes underneath them. At the start it is mostly making the thing. Then customers arrive and some of the week becomes support and sales. Then the first hires arrive and some of it becomes interviewing, onboarding and one-to-ones. Then money needs raising or managing, and a spreadsheet becomes a place you live.

Each stage adds work and rarely removes any. The description in your head lags behind, sometimes by a year. You still think of yourself as the person who builds the product, while your Mac spends its day showing you the inbox.

None of this is a failure. A company that is working generates coordination, and someone has to do it. The trouble is only that decisions about your own time get made on the old description. You turn down a meeting to “protect building time” that was never going to be spent building, or you agree to a recurring call because it is “only thirty minutes”, in a week that has no thirty minutes left.

Measuring a week without a system

Founders are bad candidates for time tracking. There is no single project to log against, the day changes every hour, and a timer is the first thing to be forgotten. So do not track. Record, and sort later.

1. Pick an ordinary week. Not the week of a board meeting, a launch or a fundraising sprint. You want the week that repeats.
2. Write your guess first. Four jobs, four numbers adding up to a hundred: building the product, selling and customers, people and hiring, running the company (money, legal, admin). Put it somewhere you will not reread until Sunday.
3. Let the Mac record which app is in front. Screen Time can do this (System Settings, then Screen Time, then App & Website Activity), though it runs through evenings and weekends and does not know where your working day ended. Punchcard sits in the menu bar, notices the front app by name only, and prints a paper receipt at the closing time you set: one line per app, a time on each, a day total. On Sunday it prints a week roll.
4. Write one line each evening. Which hat did today mostly wear? “Hiring, two interviews and a debrief.” “Customer escalation, then the pricing model.” Thirty seconds, and it is the part that makes step five possible.
5. On Sunday, lay out the week. The week roll or five receipts, plus five lines of notes.

Sorting the apps into the jobs they served

Here is the honest limit, and it is a large one for founders: Punchcard shows apps, not which part of the business an hour belonged to. Mail served all four jobs this week. A spreadsheet could be the financial model, the hiring pipeline or the list of customers who are about to churn. The browser might be the product itself, if you are building a web app, or the bank, or an hour of reading. The receipt records app names only, never window titles, document names or URLs, so it cannot tell these apart, and nothing will.

That is why the evening line exists. Sort in three passes.

1. Lines that belong to one job. The code editor or design tool is building. The accounting software is running the company. Keynote on a week you were pitching is raising money.
2. Lines that were mixed. Split them using the notes. If Tuesday was “hiring” and Mail had two hours that day, most of Tuesday’s mail was hiring. It is rough. Rough is enough.
3. Lines you cannot place. Leave them in a column called coordination, which is an answer in itself.

If the browser is your biggest line, Your browser is the biggest line on your time report covers ways to split it. The quickest for a founder is to give your own product its own app: on macOS Sonoma or later, Safari’s File, then Add to Dock turns a site into a web app, and it gets its own line on the receipt. For the narrower question of admin alone, How much of your week is admin when you run a business from a Mac goes through that sort.

Reading the gap between the guess and the week

Put your four guessed numbers next to the sorted week. A few shapes come up often enough to name.

Building is smaller than you said. The product line is the first thing to shrink as a company grows, and founders are usually the last to notice, because they still introduce themselves as builders. If the editor had four hours across the week and you guessed half your time, that is the finding.

Coordination is the biggest thing you do. Mail, chat, Calendar and a video-call app together often outweigh everything else. That is not necessarily wrong. It becomes a problem when it expands into every unprotected hour, which is what coordination does if nothing stops it. Meetings versus making is a good way to look at that split more closely.

The job only you can do is missing. Often it is selling. If nobody else at the company sells and your week had almost none of it, the week has told you something your plan did not.

Remember what the Mac cannot see. Calls taken on your phone, a coffee with a candidate, the train ride where you finally thought the pricing through: none of it appears. Punchcard has no iPhone app. Add those hours from your calendar, and if a large share of your week happens away from the Mac, read the receipts as one part of the picture.

What to do with the answer

“Spend more time on the important thing” is a resolution, not a plan. Try this instead.

1. Write down the job for this quarter in one sentence. “Close the first ten paying customers.” “Hire two engineers.” The sentence decides which lines are too big.
2. Take the largest line that does not serve that sentence and choose one of three things. Hand it off (someone else runs the weekly sync). Batch it (mail at 9 and 4, not all day). Drop it (the report you write that nobody reads).
3. Protect one block for the job only you can do. Put it in Calendar where the rest of the company can see it. A block that exists only in your head does not survive Tuesday.
4. Record another ordinary week in a month. Compare the two week rolls. If the line you meant to shrink shrank and the job for the quarter grew, the change worked. If not, you have learned where the time actually comes from, which is the next thing to change.

Do not aim for a week with no coordination. Part of running a company is being the person other people can reach. The aim is a week that looks like the job you decided to do, rather than the job that happened to you.

Questions

Does Punchcard know which part of the business I was working on?

No. It records app names and time only, with no projects, tags or clients. The same boundary is why it is comfortable to run on a Mac full of board decks, salary spreadsheets and cap tables: it never reads window titles, document names or screen contents, and it contains no networking code, so nothing about your week leaves the machine.

Can I use it to see where my team’s time goes?

No, and it is not built to. There are no team features and no account; each receipt stays on the Mac that printed it. It is for looking at your own week. A founder who wants to understand the team’s time is better off asking them.

How often should I do this?

Once a quarter is plenty, or whenever the job feels wrong and you cannot say why. Tracking never expires, so after the first week you can simply keep the receipts running and look when you need to.

What if my week was mostly away from the Mac?

Then the receipts show the desk part of the job, which may be the smaller part. Add meetings, calls and travel from Calendar, and treat the Mac record as a check on the hours you spent at the desk rather than a picture of the whole week.