Guide
Time tracking for freelancers means logging hours per client and project so you can invoice accurately and see your real hourly rate. Harvest wins if you want time and invoices in one place, with a free plan for one seat and two projects. Clockify wins for a free start across many platforms. Timing wins for automatic tracking with billing status on a Mac.
Time tracking for freelancers is the practice of recording how long you spend on each client, project and task, so you can bill correctly and see what your work actually earns. It covers the billable hours on an invoice and the unpaid hours around them, such as proposals, email, revisions and admin.
Most people get it wrong by treating it as a stopwatch for invoices and nothing else. They track only the time a client pays for and start the timer late. Then they rebuild the rest from memory on Friday and never look at the record again. That approach produces an invoice. It also hides the most useful thing the data can tell you: how much of your week never gets paid.
Disclosure: we make Dayflow, one of the tools in the table near the end. The rest of this guide is about the practice, not a product.
An employee’s hours usually feed payroll. A freelancer’s hours feed three different things, and a good tracking habit serves all three.
Billing. If you charge by the hour, the time record is the invoice. Each line needs a date, a client, a project, a duration and a note a client would accept as an explanation. If you bill a retainer, the record shows whether the client is using up the retainer or going over it.
Pricing. If you charge fixed fees, time data tells you whether those fees work. Suppose a flat-price project takes twice the planned hours. It paid half the rate you thought it did. Without a record, you quote the next project on the same guess.
Evidence. A client may ask what they paid for or dispute a scope change. A dated log of work is far stronger than a reconstruction in either case. The same log helps you write status updates and remember what you did on a project months later.
A typical time entry has a handful of fields:
| Field | Why it matters |
|---|---|
| Date and duration | The raw unit you bill or analyze |
| Client and project | Decides whose invoice the entry lands on |
| Task or note | Explains the line to the client |
| Billable or non-billable | Separates paid work from overhead |
| Rate | Converts hours to money and can differ per client or project |
| Billing status | Shows whether the entry has been invoiced yet |
If you want a starting structure, the billable hours log template uses these columns.
Four ideas get blurred together. They work in similar ways but answer different questions.
| Who it serves | Question it answers | Typical output | |
|---|---|---|---|
| Freelance time tracking | You and your clients | How many hours went to whom, and what did they earn? | Invoices, effective rate, project history |
| Timesheet | Usually a client or agency | What hours are you claiming for this period? | A dated grid of hours, sometimes approved |
| Productivity tracking | You | Where does my attention go, and what distracts me? | Focus patterns, app usage, habits |
| Employee monitoring | An employer | Is this person working during paid hours? | Activity levels, sometimes screenshots |
A timesheet is an output of time tracking, not the same thing. Some clients and agencies ask for one in their own format, so your raw record needs enough detail to fill it in. A timesheet template can bridge the two.
Productivity tracking overlaps the most. The same automatic record that shows two hours in a client’s codebase also shows forty minutes in a news app. If you use both views, you get billing data and an honest picture of your attention from one source.
Employee monitoring is the one to keep separate. It exists so someone else can check on you. As a freelancer, you own your records and you decide what a client sees. If a client asks you to install monitoring software on your machine, that is a contract question, not a time tracking one. Time tracking without surveillance covers that line in more detail.
Every approach does the same two jobs. It captures what happened, and it assigns that time to a client and project. The approaches differ in when and how each job happens.
You press start when you begin a task and stop when you finish. The entry is assigned as you create it, so it is ready to bill. The weak point is human: you forget to start, forget to stop, or leave a timer running over lunch. Timers suit people who work in long, clearly scoped blocks.
You enter hours at the end of the day or week. This is fast and flexible, but it relies on memory, and memory compresses. Short tasks disappear, and long tasks get rounded to whatever feels right. It works best on days with few switches between tasks.
Software records what you do in the background, such as which apps, documents and websites were active. You assign blocks of that record to clients afterward. Nothing depends on remembering to press a button, so small tasks show up.
The trade-off is that you still have to review and assign the record. You also need to be comfortable with what gets captured and where it is stored. Our guides to passive time tracking and automatic time trackers go deeper.
Three settings shape what the record turns into.
Yes, even though no client will see the hours. A fixed fee is a bet on how long the work will take, and time data is the only way to find out whether you won or lost it.
Tracking fixed-price work does three things:
That baseline matters because it is easy to be optimistic about how long your own work takes. The time estimation test is a short exercise that compares your sense of time with a clock, if you want to see how your guesses hold up.
This is a loop, not a tool feature. It works with a timer, a spreadsheet or automatic tracking.
Step 2 is the one people skip, and it matters most. A record assigned the same day is accurate. A record assigned at month end is a reconstruction. For more on the billing side, see how to track billable hours on a Mac.
This is an invented example, not a real person or real data.
Maya is a freelance product designer with a standard hourly rate. She has two clients:
Here is one of her weeks after she starts capturing everything:
| Work | Hours | Billable? |
|---|---|---|
| Retainer client: design and reviews | 16 | Yes, hourly |
| Website project: design | 9 | Covered by fixed fee |
| Website project: extra revisions requested by email | 3 | Covered by fixed fee |
| Email and messages, all clients | 4 | No |
| Proposal for a new lead | 3 | No |
| Invoicing, bookkeeping and admin | 3 | No |
| Total | 38 |
Three things show up that a billable-only timer would have hidden.
First, only the retainer client pays for the hours she logs. That means 16 of her 38 hours this week turned directly into invoiced time. Everything else was either fixed-fee work or overhead.
Second, the website project has already used 12 of its 30 planned hours. Three of those hours were revisions that arrived by email and were never discussed as scope. If that pattern continues, the project will run well past 30 hours, and her effective rate on it will fall below her standard rate.
Third, 10 hours went to email, a proposal and admin. That is ordinary for a freelancer, but it means her real hourly rate across the week is well below her headline rate. The real hourly rate calculator does this arithmetic for you.
After this review, she makes three changes:
If remembering to start and stop timers is the hard part, tracking billable time with ADHD covers approaches that do not depend on it.
The details below come from each vendor’s own website, as checked on October 4, 2026. Prices are in USD and can change, so confirm on the vendor’s pricing page before you buy.
| Tool | What it does for this | Price |
|---|---|---|
| Timing | Mac-only automatic tracking of apps, websites and documents. Billing status, report rounding and GrandTotal invoicing integration depend on plan (Timing vs Dayflow) | Professional $9 per month billed annually, or $11 billed monthly; Expert $11 per month billed annually, or $14 billed monthly; no free tier, 30-day trial |
| Harvest | Timers and timesheets on web, desktop and mobile, with invoicing, online payments and expense tracking | Free plan, $0 forever for 1 seat and 2 projects; Teams $9 per seat per month ($108 per seat, billed annually), or $11 billed monthly |
| Toggl (formerly Toggl Track) | Timers, timesheets and calendar integrations across platforms, with billable rates on paid plans (Toggl for Mac) | Free for up to 3 users; Starter $9 per user per month billed annually, or $12 billed monthly |
| Clockify | Timer, timesheet and a desktop auto tracker on web, Mac, Windows, Linux and mobile. Invoicing starts on the Standard plan (Clockify for Mac) | Free for up to 5 users; Basic $3.99 per seat per month billed annually, or $4.99 billed monthly; Standard $5.49 billed annually, or $6.99 billed monthly |
| Tyme | Manual timers on Mac, iPhone, iPad and Apple Watch, plus geofence clock in and out, expense and mileage tracking, and invoicing plugins | $4.99 per month per user ($59.99 annual payment), or $6.99 per month per user billed monthly; no free tier, 14-day trial |
| Dayflow | Records your screen on a Mac (macOS 14 or later) and has AI write the day into a timeline you can review and export as Markdown | $0 and MIT licensed; optional Pro plan at $20 a month |
Disclosure again: Dayflow is our product. It fits freelancers on macOS 14 or later who want a full automatic record of the day to assign to clients afterward, with recordings and the timeline stored on the Mac and screenshot analysis kept on-device if you use local AI. It does not fit if you need invoices, payroll approvals, team reports or website blocking, because it has none of those, and its AI summaries can miss or misinterpret work, so check entries before they reach a client.
For a broader comparison, see our roundup of Mac time trackers and our freelancer-specific list.
Time tracking for freelancers is less about the timer than about the habit around it:
The invoice is the obvious payoff. The bigger payoff is knowing your effective rate, which projects run over and how much of your week goes to unpaid work.
Choose a capture method that matches how you work. Timers suit long, focused blocks. After-the-fact timesheets suit quiet days with few switches. Automatic tracking suits fragmented days where small tasks would otherwise vanish. Then choose a tool by the output you need, because a tool that cannot produce that output makes the habit hard to keep through a busy month.
Pick one capture method you will actually use: timers, an end-of-day timesheet or automatic tracking. Create a project for each client plus one for admin. Record all working time for two weeks without changing anything else. At the end of each day, assign entries to clients and add short notes. After two weeks, review billable versus non-billable hours before you decide on a paid tool.
Billable hours are time a client pays for, either directly on an hourly invoice or within a retainer. Non-billable hours are everything else that keeps the business running: email, proposals, sales calls, bookkeeping and learning. Both count as work. If you track only billable hours, your headline rate looks like your real one. Your real rate is total earnings divided by all hours worked.
Timers work well for long, clearly scoped blocks and produce entries that are ready to bill. They depend on remembering to start and stop. Automatic tracking records activity in the background, so short tasks and switches are not lost. You still have to assign that record to clients afterward. If your days are fragmented, automatic capture is usually easier to keep accurate.
Yes. Harvest has a free plan for 1 seat and 2 projects that includes invoicing and expense tracking. Clockify is free for up to 5 users, though invoicing needs a paid plan. Toggl is free for up to 3 users, but billable rates need a paid plan. Dayflow is $0 and MIT licensed on macOS 14 or later, but it has no invoicing.
No. Dayflow does not create invoices. Harvest is the better choice if you want tracked time to turn into invoices with online payments, and its free plan covers 1 seat and 2 projects with invoicing included. Dayflow is useful for building an automatic record of your day that you can export as Markdown and carry into another tool. The billing step happens elsewhere.
It depends on the tool and its settings. Dayflow stores recordings and the timeline on your Mac. With local AI through Ollama or LM Studio, screenshot analysis stays on-device. With your own AI provider or Dayflow Pro, selected screenshots are processed off-device, and under its privacy policy Pro deletes screenshots after processing. Timing says tracked times stay on the Mac unless you opt into sync.
Rebuild the gap from evidence rather than memory. Your calendar, sent email, file edit times, commits and messages are all useful. Enter the reconstructed time with a note that it was added later, and be conservative when you bill it. Then remove the cause. If you often forget to start timers, an automatic record that you assign afterward removes that step entirely.
Share what the contract requires and what supports the invoice: dates, durations, projects and clear notes. You do not need to share raw activity data, app usage or anything else a tool captured beyond the work itself. Your record belongs to you. If a client asks you to run monitoring software on your own machine, treat it as a contract negotiation rather than a default.
A spreadsheet is enough if you have few clients, work in long blocks and fill it in every day. A dedicated tool helps when you need timers, rates per client, billing status or invoices. It also helps if you keep forgetting to log short tasks. Start with a template, and move to software once the spreadsheet starts costing you time or accuracy.
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