billable hoursfreelancingautomatic time trackingguide

How to track billable hours without starting a timer

If you bill by the hour and keep forgetting to hit start, the problem isn't discipline — it's the timer. Here's how to capture billable hours automatically from the work you're already doing, and turn them into invoices without reconstructing your week from memory.

3 min read
A stopwatch crossed out — track billable hours without starting a timer

The real reason you keep forgetting the timer

Every freelancer who bills by the hour has lived this: it's Friday afternoon,
the invoice is due, and you're staring at git log, your browser history, and
a half-remembered week trying to figure out how long the Acme work actually
took. You round. You guess. You almost certainly under-bill.

The usual advice is "build the habit." But a manual timer asks you to do
something — start, switch, stop — at exactly the moments your attention is
already gone: a context switch into a different client's repo, a quick jump to
their staging site, a Slack thread that turns into an hour of real work. The
timer loses that fight every time. It's not a discipline problem. It's the
wrong tool for how the work actually happens.

The alternative: capture, don't start

Instead of asking you to mark when work begins, automatic time tracking
watches the work itself and infers the session.

Two signals cover most billable work:

  • File edits. A background desktop client (macOS, Windows, Linux) watches
    the directories you tell it to — say ~/code/clients/acme/. The moment you
    save a file there, a session for that project starts. On a headless box or
    VPS, the Linux CLI does the same over SSH.
  • Visited URLs. Browser extensions (Chrome, Firefox, Edge, Safari) map URL
    patterns to projects. Time on github.com/acme-co/* or the client's staging
    domain lands on the Acme project automatically.

You set the paths and patterns once. After that, the question "did I start the
timer?" stops existing — there's nothing to start.

A file edit and a visited URL flowing automatically into a running session tagged to the Acme project, started via the watcher with no timer

What about the breaks?

The objection to automatic tracking is always the same: won't it bill my lunch?
No. When you stop touching in-scope files for a configurable idle window
(default 15 minutes), the session auto-closes. The gap isn't billed. When you
come back and edit again, a fresh session opens. If you want to split the day
deliberately, a pause/resume control is one click.

From captured time to a paid invoice

Tracking is only half the Friday-afternoon problem; the other half is turning
hours into a bill. Because projects link to clients and rates, a captured
session becomes an invoice line item directly — no export to a separate
invoicing app, no copy-paste. You review the week on the dashboard, fix
anything that looks off, and generate the invoice from the same time you just
tracked automatically.

Captured Acme sessions for the week totalling 7h 40m turning, in two clicks, into an invoice line item billed at an hourly rate

What this approach deliberately does not do

Worth being clear, because "automatic tracking" means surveillance in some
products and it doesn't here:

  • No screenshots, ever. Capture is file paths and URLs — both auditable in
    plain text.
  • No keystroke logging and no productivity score. It records what you
    worked on and where you browsed, not a judgement about it.
  • No reconstruction from memory. The session was recorded when it happened.

How to switch without losing past data

If you're moving off a manual timer, CSV import brings your historical entries
in, so the new automatic record sits alongside what you've already billed.

The short version

You don't forget to start a timer because you're undisciplined — you forget
because the timer interrupts the work. Capture the work instead: watch the
files you edit and the URLs you visit, let idle time close sessions, and
invoice from what was captured. Friday afternoon stops being archaeology.

Frequently asked questions

Won't automatic tracking bill time I wasn't really working?

No. Sessions close automatically after a configurable idle window (default 15 minutes) with no file activity, so breaks aren't billed. You can also pause and resume manually to split a day.

How does it know which client a block of time belongs to?

You map directories and URL patterns to projects once. Editing files under a project's watch path, or visiting a matching URL, tags that time to the right project automatically.

Do I still get a manual timer if I want one?

Yes. A manual timer is available as a fallback, but the default expectation is that you never touch it — tracking happens on its own from your file edits and visited URLs.

Can I turn the tracked hours into an invoice?

Yes. Projects link to clients and rates, so a captured session becomes an invoice line item in the same tool — no export to a separate invoicing app.

What about the hours I've already logged elsewhere?

CSV import brings historical entries in so your past billable time sits alongside the automatically captured sessions.