Guide
Measure three things first: the hours you work now, what a finished week looks like, and how often work leaks into the day off. A paper log and your calendar handle all three and win if you want no software. Dayflow, a free Mac app, wins if you won't remember to log, since it records your screen and writes the timeline for you.
You take a Friday off to see what a four-day week feels like. By ten you’ve answered two emails “just to get ahead of Monday.” By two you’re on a call you didn’t feel able to decline. On Monday you can’t say whether the experiment worked. The week felt both shorter and more crowded.
That’s the usual first attempt when nobody else is on the schedule. No company policy protects the day, no team has moved its meetings, and nobody has agreed how to tell whether the work still got done. So the real question isn’t “can I work four days?” It’s “what would I need to see to know this is working?”
This guide covers what to measure before and during a solo trial: your current hours, what a finished week looks like, how often work leaks into the day off, how long the other days get, how much energy is left, your meetings, and money if you’re paid by the project. Most of it needs a notebook and a calendar. One section covers keeping an automatic record of your screen. Dayflow is our product, and that section is about it.
Mostly because the other people in your week are still on five days. Messages keep arriving on your day off, and replying feels polite and small. Each reply is small. Together they turn the day off into a lighter workday.
Work also expands to fill whatever room it’s given, and it shrinks less easily than it grows. If you don’t decide what a week’s work is, the four days fill to bursting and the fifth picks up the overflow.
Memory makes it worse. At the end of a week most people remember the deadlines and the stressful calls, not the hour spent tidying a document that didn’t need it. If you judge the trial from memory, you’re judging it from the loudest moments. (If you often finish a day unsure where it went, where did my day go covers that problem on its own.)
Finally, a solo trial has no agreed finish line. A team pilot usually has a start date, an end date and something to report. On your own, one bad Thursday can feel like proof that it can’t work, and you quit after a week that told you very little.
None of this is a willpower problem. It’s a measurement problem, and that’s the fixable part.
What to do. Before you drop a day, spend two ordinary weeks writing down what you work on, roughly each hour. A sheet of paper by the keyboard is enough. Each time you notice the hour has turned, write one line: the time and what you were mostly doing. Gaps are fine. Mark them as gaps instead of guessing later. A time audit template gives you a ready-made sheet, and the where does my week go calculator gives you a rough first picture if you want one before you start logging.
An illustration. Someone who believes they work “about forty hours” logs two weeks and finds most days start with half an hour of reading messages that she hadn’t counted as work, and end with a late check of email she hadn’t counted either. Her actual total is higher than she thought, and much of it is scattered.
How to tell after a week. You should be able to answer three questions without guessing: roughly how many hours you worked, which day was longest, and which kind of task took more time than you expected. If you can’t, log a second week before changing anything. Comparing against a fuzzy baseline is how trials get judged on mood.
What to do. Write five to eight lines describing a normal, finished week of your work. Use outcomes you can check, not effort. “Sent the client draft.” “Closed the support queue.” “Reviewed two pull requests.” Keep the list somewhere you’ll see it on your last working day, and tick it off then. A weekly status report template works well for this, even if nobody else reads it.
An illustration. A designer’s list might read: two rounds of feedback answered, one new screen drafted, the weekly handoff call held, invoices sent. In week one of the trial, everything gets ticked except the new screen. In week two, the screen slips again. That repeat is the useful signal. Creative, uninterrupted work is the first thing squeezed out of a shorter week, so that’s where to protect time.
How to tell after a week. If the list got done in four days, the extra day isn’t costing you much. If one item keeps slipping, you know where the pressure sits. If the whole list is half done, the trial needs adjusting, perhaps by dropping a weekly task or moving the day off. You don’t have to give up on it.
What to do. Put a sticky note somewhere you’ll see it on your day off. Each time work reaches you, make a mark and a two-word note on the source: “email, phone,” “chat, client,” “thought, shower.” The night before the day off, set an out-of-office reply, block the day on any calendar others can see, and sign out of work chat on your phone.
An illustration. After the first off-day, the note shows five marks. Three came from work chat on the phone, one from email, and one was just remembering a task while cooking. Signing out of chat on the phone takes care of most of it. The remembered task goes on a “Monday” list kept by the bed, so it can be dropped without being done.
How to tell after a week. Count the marks. You aren’t aiming for zero on day one. You’re aiming for the number to fall and for the sources to change. If the same source shows up every week, that door needs a firmer close: a clear note to clients about your hours, or moving the day off to one where fewer people expect a reply.
What to do. For each working day, write down when you actually started and actually stopped. Use the first and last moment of real work, not the time you sat down. Your calendar is a fine place for this. Add a block called “worked” from start to finish after the fact.
An illustration. Week one: Monday to Thursday run from about nine to six. Week three: start times have crept to eight-fifteen and Thursday ran until eight at night. The day off survived. The hours just moved into the other four days without any decision being made.
How to tell after a week. Compare the length of your working days against your baseline. If you chose to compress the same hours into four days, longer days are expected, but check they stay at a level you’d be happy keeping for months. If you meant to work fewer hours overall, longer days mean the work hasn’t shrunk, and it’s time to revisit the finished-week list.
What to do. Pick a fixed time each evening and write one number from one to five: how much attention and energy you have left for the rest of your life tonight. One means nothing, five means plenty. Add one line on why if something stands out. Do this during the baseline weeks too. An end-of-day report template has room for the number and the line.
An illustration. Baseline weeks hover around two and three. In the trial, Mondays drop to two because of the backlog, but Wednesday and Thursday sit at three and four, and the day off is a steady five. Over time that pattern says more than any single week.
How to tell after a week. Look at the average, then the lowest day. A higher average with one brutal day may be a fine trade. Lower numbers across the board suggest the compressed days are costing more than the free day gives back. Either way, you’re looking at your own notes instead of a vague sense of being tired.
What to do. List every recurring meeting on your calendar and mark each one: keep, shorten, make async, or drop. Look hard at the ones on your chosen day off. Ask whether those can move, rather than whether you can skip them, because skipping tends to invite catch-up calls later. The meeting cost calculator can help you frame what a recurring hour costs when you raise it with others.
An illustration. A developer finds a weekly sync on Fridays, a status call that’s mostly reading updates aloud, and two one-to-ones. The Friday sync moves to Thursday afternoon. The status call becomes a written update in the team channel. The one-to-ones stay.
How to tell after a week. Count meeting hours against the baseline. If they dropped and nothing broke, keep the changes. If a meeting you made async keeps turning into a long message thread, it may need to come back in a shorter form. And if the day off keeps getting meetings booked onto it, the calendar block isn’t visible enough to others.
What to do. This one matters most if you freelance or bill by the project. Note what you earned over the baseline weeks and the hours from your log, then do the same over the trial. Fixed-fee work can mean four days earn the same as five. Hourly work usually means less income unless your rate changes. The real hourly rate calculator helps you include the unbilled time, like admin and email, that people tend to forget.
An illustration. A consultant on fixed monthly retainers finds her income unchanged across the trial while her logged hours fell, so her effective rate rose. A colleague who bills hourly sees income fall in step with hours and decides to raise his rate before trying again.
How to tell after a week. One week of income is usually too noisy to read. Note it anyway and judge it across the whole trial. The useful question is whether the money works at the hours you want, not whether a given week was good.
Every step above depends on you remembering to write things down, and that’s what slips first on a busy day. If you work on a Mac, an automatic record takes that job off your hands.
Dayflow records your screen at one frame every 10 seconds and has AI write the day into a readable timeline. On Thursday evening you can see when work actually started, how long the afternoon call ran, and whether your day off included a quiet visit to the inbox. That covers the hours, day length and leak measures without you starting a timer.
Recordings and the timeline are stored on your Mac. With Local AI, a model runs on your Mac through Ollama or LM Studio and screenshot analysis stays on-device. You can also send selected screenshots to a cloud provider you configure, or use Dayflow Pro at $20 a month, which deletes screenshots after processing under the privacy policy. Local and cloud processing explains the difference. You can export a date range as Markdown from Settings → Export to compare baseline weeks with trial weeks. AI summaries can miss or misread work, so check the timeline before relying on it.
Dayflow is $0, MIT licensed and needs macOS 14 or later. It’s one of several Mac time trackers.
Dayflow is our product and Dayflow publishes this article. Details here come from Dayflow’s published documentation for version 2.2.0, reviewed on 2026-09-06.
It’s Mac only. If much of your week happens on a phone, a Windows machine, in a workshop or in meeting rooms, Dayflow won’t see it, and a paper log will give you a more complete picture.
It sees only the screen. Thinking on a walk, a phone call on your mobile and a whiteboard session don’t show up. The finished-week list and the evening energy number have to come from you either way.
It’s a record, not a blocker. Dayflow has no website blocking, so it won’t keep you out of email on your day off. An out-of-office reply and signing out of mail on your phone do that job better.
It has no invoicing, payroll approvals or team reports. If your employer is running a formal pilot and wants numbers from its own system, use that system.
It doesn’t tell you whether the trial is working. That judgment is yours, made against the list and the baseline.
And if being recorded at all feels like too much during an experiment meant to give you breathing room, use a notebook and a kitchen timer. That’s enough for everything in this guide.
A solo four-day week rarely fails because you can’t do the work in four days. It fails because the fifth day gets nibbled away, the other four quietly grow, and nobody wrote down what a good week looked like. Measure your baseline first, write the finished-week list, tally the leaks, and pick a trial length before you start. A notebook and a calendar will carry you most of the way. If you’d rather not rely on remembering to log, an automatic record on your Mac can fill in the hours, but the decision about whether the shorter week works for you still comes from your own notes.
Start with a personal trial on whatever room you already have. That might be a protected afternoon, a no-meetings day, or a day off you've saved leave for. Measure your normal week first, then run the shorter one for a few weeks with the same measures. If you later ask your manager for something formal, you can bring a record of what got done instead of a feeling.
Keep it small. Track the hours you actually work each day, a short list of what counts as a finished week, how many times work reaches into your day off, and one number for how much energy is left each evening. Record the same things during a normal baseline week so you have something honest to compare against.
A single week tells you very little, because one deadline or one quiet week can swing everything. Many people find that a two-week baseline followed by about a month of the shorter week gives a fairer picture. Decide the length before you start and write it down, so a single bad Thursday doesn't end the experiment early.
That is one version, often called a compressed week, and it suits some people. It's a different experiment from working fewer total hours. Pick one on purpose. If you compress, watch whether the four days stretch past what you can sustain. If you reduce hours, watch your output list closely, since that is where a shortfall would show first.
Make the day visible to others and harder to reach. Block it on your shared calendar, set an out-of-office reply the night before, and sign out of work chat and email on your phone. Then keep a tally of every time work gets in anyway. The tally tells you which door is still open, which is more useful than willpower.
Write a short list of what a normal finished week contains before you start: things shipped, clients answered, drafts sent. Check each shorter week against that list on the last working day. If the list keeps getting done, the extra day isn't costing much. If the same item slips every week, that tells you where the time pressure sits.
No. Dayflow has no website blocking, so it will not stop you opening your inbox. An out-of-office reply, signing out of mail on your phone, or a dedicated blocking app does that job better. Dayflow can only show you afterward that the inbox was open on your day off, which is useful for the tally but does not prevent it.
No. Dayflow runs on macOS 14 or later and only sees what is on your Mac's screen. Work done on a phone, a Windows machine, or in person won't appear in its timeline. If much of your week happens away from a Mac, a paper log or a simple calendar record will capture more of it.
Yes. Dayflow is $0 and MIT licensed, with an optional Pro plan at $20 a month where Dayflow manages the cloud analysis. On the free app you can run a local model through Ollama or LM Studio, which keeps screenshot analysis on your Mac, or connect your own cloud provider, whose charges and privacy terms then apply.
Dayflow is $0 and open source. Full recordings stay on your Mac, with local AI available for on-device analysis.
Download for MacOS macOS 14+ · MIT licensed · no account required