How to Do a Time Audit (The Freelancer Method)
A time audit is a week of honest measurement: you record where every working hour actually goes, sort it into categories, and compare the result against where you assumed it went. For freelancers the gap is usually uncomfortable and always useful, because your rate, your capacity and your pricing all rest on a number most people have never measured.
The point is not self-discipline. It is a baseline. You cannot decide whether to raise your rate, drop a client or hire help until you know what your week is really made of.
What is a time audit?
A time audit is a fixed-length measurement of how your working time is distributed, run once, on purpose. It differs from ordinary time tracking in two ways: it covers everything, not just billable client work, and it has an end date.
You are trying to answer three questions:
- How many hours did I actually work?
- How many of those hours were billable?
- Where did the rest go?
Most freelancers can guess question one, are wrong about question two, and have no answer at all for question three. Question three is where the money is.
How do you run a week-long time audit?
One ordinary week. Not your quietest, not your worst. Here is the method.
Day 0: set your categories before you start. Deciding categories mid-week corrupts the data. Five or six is plenty:
- Client work, billable (by client)
- Client communication (email, calls, Slack)
- Revisions and rework
- Business admin (invoicing, chasing payment, bookkeeping)
- Sales and proposals
- Internal work (your own site, learning, tooling)
Days 1 to 5: record at the point of work, not at the end of the day. This is the part that decides whether the audit is worth anything. Retrospective reconstruction reliably loses the small stuff, and the small stuff is the finding. If you are logging by hand, write the entry when you switch tasks, not when you remember.
Include the edges. The 7am email triage, the call you took walking, the twenty minutes after dinner. Excluding them because they were "not really work" is how the audit tells you a comfortable lie.
Day 6: total it. Hours per category, hours per client, and your billable share of the whole.
Day 7: run one week more if the week was strange. A single week that happened to contain a launch or a dead spell is a story, not a baseline. Two weeks is enough for most people.
If a week of manual note-taking sounds like exactly the discipline problem you were trying to solve, that is a fair objection. It is the reason passive time tracking exists: the computer records the active window in the background and you categorise afterwards, which turns the audit into a review job instead of a memory job.
A simple time audit template
If you are recording by hand, keep the sheet narrow enough that filling a row takes seconds. Five columns is enough:
| Time block | Activity | Client | Billable? | Notes |
|---|---|---|---|---|
| 09:00 to 09:40 | Inbox and replies | Mixed | Partly | Two client threads, one admin |
| 09:40 to 11:15 | Homepage build | Acme | Yes | Uninterrupted |
| 11:15 to 11:30 | Status call | Acme | No | Outside scope, unbilled |
| 11:30 to 12:10 | Chasing an unpaid invoice | - | No | Overhead |
Copy it into a spreadsheet with one tab per day. The Notes column is the one people drop first and the one that makes the totals readable a week later: "Partly" with no note beside it is a row you cannot categorise on day six.
How should you categorize what you find?
Once you have the raw hours, sort them along two axes. Most people only use the first one, and the second is where decisions come from.
Axis one: billable or not. Straightforward, and covered properly in billable vs non-billable hours.
Axis two: does this scale with revenue, or is it fixed overhead? Client work scales: more clients, more hours. Bookkeeping mostly does not. Proposal writing sits awkwardly in between. This axis tells you what to fix and what to price in:
| Category | Typical share of a working week | What to do with it |
|---|---|---|
| Billable client work | The number your rate depends on | Protect it |
| Client communication | Often larger than expected | Decide what is billable, in writing, per client |
| Revisions and rework | Grows quietly on fixed-fee work | Cap it in the agreement |
| Admin and invoicing | Fixed overhead | Automate or batch it |
| Sales and proposals | Lumpy, unavoidable | Price it into your rate |
The shares are deliberately left blank. Anyone who tells you the correct percentages for a freelance week is guessing, and the honest ones say so. We went through which of these widely quoted figures survive scrutiny in time tracking statistics, and most do not.
What does a time audit usually reveal?
Three findings come up again and again. Yours may differ, which is the point of measuring rather than reading.
Admin is bigger than you think. Invoicing, chasing late payment, expense sorting and inbox management often consume the equivalent of a full working day a week. It is genuinely necessary work and it is genuinely unpaid, which means it belongs in your rate calculation rather than in your resentment.
Your billable share is lower than your assumption. Almost nobody bills 40 hours out of a 40-hour week, and nobody should expect to. What matters is knowing your actual figure so you can price against it rather than against a fantasy. Billable utilization rate has the formula, and how many billable hours per week is realistic has the benchmarks.
One client is quietly more expensive than the rest. Not the one you dread. Usually the pleasant one who sends lots of small requests. When you total the calls, the "quick questions" and the extra rounds, their effective hourly rate can be half of what their contract says. That is scope creep, and it only becomes visible when the hours are attributed by client.
The fourth finding is harder to see in a single week: the tax you pay for jumping between clients all day. It shows up as short, fragmented blocks that never add up to a full task, and it deserves its own measurement.
How do you turn the audit into a decision?
An audit that ends in awareness was a waste of a week. Each finding maps to a specific decision.
- Low billable share, full calendar. You are busy but underpaid. Raise your rate or cut the overhead, and use your measured billable hours (not 2,080) as the divisor when you recalculate: how much to charge as a freelancer walks through the arithmetic.
- High admin load. Batch it into one block a week, or remove the manual step entirely. Invoicing is the usual culprit and the easiest to automate.
- One client eating disproportionate time. Re-price at renewal, cap revisions, or move them to a retainer that reflects reality.
- Billable work that never reached an invoice. This is not a capacity problem, it is money you already earned and did not collect. Stop losing billable hours covers the reconciliation.
- Capacity genuinely full at a rate that works. Congratulations: the decision is whether to raise prices or find help, not whether to work more evenings.
Re-run the audit in six months, not next week. The value is in the comparison.
Making the audit continuous
The one-week audit is a snapshot, and snapshots go stale. The version that keeps paying is the one that runs by itself: capture every hour automatically, categorise it into client projects, and check the billable-versus-worked gap before each invoice rather than once a year.
That is what BillNotch does. The desktop app records the active window with no timer to start, keyword rules (plus optional AI for ambiguous activity) sort it into projects, and the Revenue Leak Finder surfaces tracked time that never made it onto an invoice. Hours become a PDF invoice or a CSV export from the same place. Pricing is flat: Pro at $9/month for one seat, Team at $12 per seat, with a 14-day trial and no card. It is not the cheapest tracker in the category; the case for it is that capture, categorisation and invoicing sit in one tool rather than three.
Where to go next depends on what your week showed. If the surprise was your billable share, start with billable utilization rate. If it was the amount of unpaid client work, read billable vs non-billable hours and settle the grey areas with your clients in writing. If you run a firm, the busy-season version of this audit is time tracking software for accountants, which records the hours you would otherwise write down.
Frequently asked questions
What is a time audit?
A time audit is a short period, usually a week, where you record everything you actually work on and then read the result. The goal is not to bill it but to see where the hours really go: which clients, which tasks, how much is billable, and how much unpaid work you had not noticed. It turns a vague sense of being busy into numbers.
How do you do a time audit?
Pick a normal week and log every work block as you go, including small tasks like calls, email, and context switches. Tag each block by client and by whether it is billable. At the end of the week, total the categories and look for the surprises. Doing it from memory defeats the point, so capture as you work or track automatically.
How long should a time audit last?
One representative week is usually enough to reveal the pattern, and two weeks smooths out an unusual one. Avoid auditing a week you already know is atypical, like a launch or a holiday stretch. If you want the picture to stay current instead of going stale, run capture continuously rather than repeating a manual audit every quarter.
What does a time audit usually reveal?
Most people find their billable share is lower than they assumed, that admin and unpaid client work eat more of the week than expected, and that a specific client or task type is quietly unprofitable. The exact percentages vary by person, so anyone quoting a single correct breakdown is guessing. Your own numbers are the only ones that matter.