A freelance designer quotes $65 an hour, bills 30 hours a month to one retainer client, and expects roughly $1,950. At tax time the real number, once every hour of email, revision rounds, and “quick call” is counted, works out closer to $40 an hour for the same work. Nothing about the invoice was wrong. The $65 rate was only ever being charged against the hours that got logged as billable, and a lot of real work never made it onto that list because nobody was tracking it.
In my experience, freelancers who avoid time tracking aren’t undisciplined about it — most tried a tool once, found it demanded a stopwatch mentality (start a timer for every five-minute task, tag it, stop it, repeat forty times a day), gave up within a week, and concluded that tracking itself doesn’t work for how they work. The real problem usually isn’t the discipline. It’s that the first tool they tried was built for a different job than the one they needed done, and nobody told them there were three genuinely different categories of tool to choose from before picking one.
This guide covers what to actually track and why it changes your real numbers, not just your invoice; an honest comparison of the three tools freelancers reach for most, based on what each one is actually built to do rather than a feature checklist; and the habit that makes tracking survive past week one. It assumes you already know roughly how you price your work — if you haven’t settled that yet, how to set your freelance rates covers the hourly-vs-project decision this guide builds on.
Your Real Rate Isn’t Your Quoted Rate
The number on a rate card is a billable rate — what gets charged for hours that are logged as chargeable to a specific client. The number that actually determines income is the effective hourly rate: total income divided by total hours worked, including everything that never gets billed to anyone. The gap between the two is usually bigger than freelancers expect, because non-billable work — proposals, admin, bookkeeping, client emails outside a project’s scope, the unpaid half of a discovery call — doesn’t feel like “real” work time in the moment, even though it fills real hours.
A commonly cited industry range puts freelancers who don’t track their time at a 50-60% billable ratio without realizing it, versus 65-75% for freelancers who actively manage the split — worth treating as a directional benchmark rather than a precise figure, since none of the vendors publishing it disclose their sample. The mechanism it points at is real regardless of the exact number: at a 60% billable ratio, quoting $65/hour and working a 35-hour week means only 21 of those hours are actually billed. $1,365 a week, not the $2,275 the quoted rate implies. Nothing changes about what was charged per hour — what changes is how many of the hours worked ever became billable in the first place, and that number is invisible until it’s tracked.
What to Track, and What Not To
Track at the project level, not the task level, for anything client-facing. A timer running against “Acme Corp — homepage redesign” for a three-hour block is useful data. A timer running against “adjust hero button padding” for four minutes is the exact granularity that makes people abandon tracking by day three — the overhead of starting and stopping outweighs anything the data tells you back. Reserve task-level detail for the rare case where a client specifically asks for an itemized breakdown, not as the default mode.
Track non-billable time too, in a separate bucket, even though nobody’s paying for it. “Admin,” “proposals,” and “business development” as three broad categories is enough — the point isn’t precision, it’s visibility into the ratio described above. Without that bucket, non-billable time doesn’t disappear, it just becomes invisible in the one place — the effective-rate calculation — where it matters most.
Three Tools, Compared by What They’re Actually Built For
All three below do the core job — start a timer, stop it, generate a report — competently. The real difference is which job each one was designed around first, and that’s what should decide which one fits a given freelancer’s workflow rather than a feature-count comparison.
- Toggl Track is built around fast, frictionless timing and reporting first. The free plan (capped at a small number of users in 2026) covers unlimited time entries, unlimited projects and clients, an auto-tracker, idle detection, and a Pomodoro timer — genuinely usable for a solo freelancer with no paid upgrade required. Billable rates and project time estimates are gated behind the Starter tier ($9/user/month), with profitability tracking and timesheet approvals reserved for Premium ($18/user/month, discounted in year one). It fits freelancers who want the lowest-friction timer and don’t need invoicing built in.
- Clockify undercuts Toggl on price at every tier and is the most generous free plan of the three: unlimited time entries and projects at $0, capped at five users. Its Basic paid tier starts at roughly $4/user/month annually, with GPS tracking, scheduling, and labor-cost budgeting reserved for the Pro tier ($7.99/user/month). Clockify fits a freelancer on a tight budget or just starting out, with the caveat that invoicing and approval workflows sit behind the same paid tiers as the advanced reporting.
- Harvest is the odd one out in a useful way: it’s an invoicing tool with time tracking built in, not a time tracker with invoicing bolted on. The free plan is limited to one seat and two projects, but even at that tier it includes real invoice generation and expense tracking — a tracked hour can become a sent invoice without leaving the app. Paid plans start at $9/seat/month for unlimited seats and add Stripe, QuickBooks, Xero, and Deel integrations. Harvest fits a freelancer who wants tracked time and the invoice it produces to live in the same system, rather than exporting a timesheet into a separate invoicing tool every month.
| Tool | Best for | Entry price |
|---|---|---|
| Toggl Track | Fastest timer UI, strongest free tier for solo use | Free, then $9/user/month |
| Clockify | Lowest cost at every paid tier, generous free plan | Free, then ~$4/user/month |
| Harvest | Tracking and invoicing in one system | Free (1 seat), then $9/seat/month |
The Adjacent Tool: When Automatic Tracking Is Worth the Extra Cost
Toggl, Clockify, and Harvest all rely on remembering to start and stop a timer. RescueTime takes the opposite approach — it runs in the background and automatically logs time by application and website, no manual start/stop required. Historically that meant personal-productivity insight only, useful for spotting how much of a day actually went to email versus design work, but not built for client billing. That’s changed: RescueTime’s current plans split the two use cases explicitly — a Solo Focus tier (from $7/month) for personal productivity insight and distraction tracking, and a separate Solo+ tier (from $12/month) that bundles the same automatic tracking with client, project, and billable-rate features. It’s a genuinely different mechanism from the other three — worth it specifically for a freelancer who knows they forget to start a manual timer more often than not, less useful for anyone who already has that habit down.
Making It Stick
- Attach the timer to an existing habit, not a separate decision. Start it as the first action when opening the client’s project file or Slack channel, the same way opening the file itself is already automatic — not as a thing to remember to do afterward.
- Review weekly, not daily. A daily check-in turns tracking into a chore; a five-minute Friday review of the week’s timesheet catches missed entries and shows the billable ratio without requiring constant attention.
- Round to the entry, not the minute. Fifteen-minute rounding on manual entries removes the temptation to fuss over exact stop times, which is usually what turns a two-second action into a thirty-second one and makes people stop bothering.
- Reconcile against the invoice, not the other way around. Generate the invoice from the tracked time, don’t estimate the invoice and check the timer after — the whole point of tracking is that the invoice reflects what actually happened.
Common Mistakes
- Tracking at task-level granularity from day one. This is the single most common reason freelancers try time tracking and quit within two weeks — the overhead of logging every micro-task exceeds any benefit the data provides at that resolution.
- Only tracking billable hours. Without a non-billable bucket, the effective-rate math above is impossible to run, which means the actual reason income feels lower than the rate card suggests stays invisible.
- Picking a tool based on feature count instead of workflow fit. Harvest’s invoicing doesn’t help a freelancer who already has an invoicing system they like; Toggl’s speed doesn’t matter to someone who wants tracked time to become an invoice automatically. Match the tool to the actual bottleneck, not the longest feature list.
- Leaving a timer running overnight. Every tool above has idle detection to catch this, but it has to actually be turned on — an eight-hour phantom entry skews a week’s numbers badly enough to make the whole exercise look untrustworthy.
A Worked Example: Auditing a Month of Retainer Work
A freelancer runs a $400/month maintenance retainer, priced on the assumption that it covers roughly six hours of work. After a month of actually tracking time against it in Clockify, the timesheet shows 9.5 hours: the six hours of expected maintenance tasks, plus 2.5 hours of “quick fix” requests that came in over email outside the retainer’s defined scope, plus one hour of a monthly check-in call that was never accounted for when the retainer was priced. Nothing in the client relationship was unreasonable — the requests were small and felt easy to just handle — but without tracked data, the retainer was quietly running at roughly $42/hour instead of the $67/hour it was priced to deliver. That gap is exactly the kind of drift how to manage scope creep covers from the client-conversation side; time tracking is what makes the drift visible in the first place, before it’s a conversation worth having.
Where This Connects
Tracked time only matters if it turns into an accurate invoice — how to write a freelance invoice that gets paid faster covers the structure that gets a correctly itemized bill paid quickly rather than sitting unopened. If retainer work is part of the mix, how to price website maintenance retainers walks through the same time-cost math this guide’s worked example runs, applied to setting the retainer price in the first place rather than auditing it after the fact. And once a month or two of real tracked data exists, it’s worth revisiting the original rate decision — how to price web design services covers calculating a minimum viable rate from actual billable hours, which is a very different number once it’s measured instead of assumed.
The Number That Actually Matters
A rate card is a target, not a measurement. The only way to know whether a freelance business is actually hitting that target is to track enough real hours — billable and non-billable both — to calculate the effective rate honestly, then adjust either the quoted rate or the non-billable overhead until the two numbers are close enough to live with. None of the three tools above fix that gap by existing. What they do is make the gap visible, which is the only precondition for closing it.
Etienne Basson works with website systems, SEO-driven site architecture, and technical implementation. He writes practical guides on building, structuring, and optimizing websites for long-term growth.