How long a video edit actually takes, measured on a Mac
· 7 min read
Nobody quotes a video edit from data. They quote it from the memory of the last one, which has quietly dropped the ingest, the three rounds of notes, the music search that ate a Thursday and the two exports that failed. Measuring a single project end to end fixes that, and it takes no extra effort during the edit itself.
The method below covers one project from first import to delivered file. Run it once and the next quote stops being a guess.
Why the estimate is always low
Editing has a shape that defeats memory. The interesting part, cutting the timeline, is the part you remember, and it is rarely the largest part. Around it sit hours that leave no trace: copying cards, transcoding, hunting for a track, waiting for a client to open a link, re-exporting because the aspect ratio was wrong for one platform.
There is also the revision problem. Round one feels like the project. Rounds two and three feel like tidying, so they get filed under “finishing” and are missing from the number you carry into the next conversation. Two afternoons of small changes is not a rounding error at a day rate.
The general version of this gap is well covered in Why your estimate of your own workday is wrong by hours. Editing is a hard case because so much of the work is waiting, and waiting is exactly what memory discards.
What to measure, and what to leave out
Decide the boundaries before you start, or the number will not compare to anything later.
Count as edit time: ingest and organizing, syncing, the cut, color, sound, titles and graphics, music search and licensing, export setup, and every round of notes including the ones that arrive as voice messages.
Count separately: the shoot, travel, client calls, scoping and the proposal. These belong in a project total but not in the “how long does the edit take” number, because they scale differently.
Do not count as your time: unattended renders and uploads. A machine exporting for forty minutes while you eat is not forty minutes of labor, although it is forty minutes of calendar. Keep the two apart or your hourly figure will lie to you in both directions.
The last one deserves emphasis on a Mac, because a record of which app is frontmost cannot tell the difference between you working in the editor and the editor sitting in front of you rendering. You will need to mark those stretches yourself, and it is one note per export.
Running the measurement over one project
- Pick a typical project, not a showpiece. A representative two-minute cut teaches you more than the one that won something.
- Let the day record itself. You want time per app for every working day of the project, collected without starting a timer, because timers on an edit day get forgotten inside the first hour.
- Keep a one-line project log. Date, phase, and anything unattended. “Tue, round two notes, 25 min export while at lunch.” Ten seconds a day.
- Mark the phase boundaries. Ingest ends when the first cut starts. Round one ends when the link goes out. This is the part that makes the total useful rather than merely large.
- Add it up on delivery day. Total hours, hours per phase, and the ratio of hours to finished minutes.
For step two the requirement is something that runs without asking anything of you. Punchcard notices which app is in front through the day and prints a receipt at the closing time you set, itemized by app with a day total. There are no projects, no tags and no timers in it, which is why it survives an edit week: nothing to start, nothing to forget, and the day is on paper before you close the lid.
The absence of project tagging is the honest limit. The record tells you the day was nine hours with six in your editor. It cannot tell you which client that was for. Your one-line log supplies what the app deliberately does not, and for a freelancer running one or two projects at a time that combination is enough. If you routinely juggle six clients in a day, you want something built for per-project billing instead.
What the numbers usually show
Three findings come up again and again once people measure a full project.
The cut is roughly half. Most editors expect the timeline to be eighty percent of the work and find it closer to half, with ingest, audio, graphics and revisions splitting the rest. That single fact reprices a lot of quotes.
Revisions are bigger than remembered and lumpier than expected. They also stretch the calendar far more than the clock, because each round includes waiting. Waiting is not billable time, but it is unavailable time, and a day with two hours of edit and a client reply pending is not a day you can sell elsewhere.
The ratio of hours to finished minutes is stable per format. A talking-head cut, a wedding film and a product spot each have their own ratio, and yours will be different from anyone else’s. Once you have two or three projects measured, this is the number you quote from. How to estimate hours for a quote when you never tracked the last job covers building a quote when you are starting from nothing.
Turning it into a price
Once you have a real total, the arithmetic is quick and slightly uncomfortable.
- Hours per finished minute, by format. Divide total edit hours by finished runtime.
- Effective hourly rate on the last project. Project fee divided by real hours, unattended renders excluded.
- Compare that to the rate you tell people you charge. The gap is the revision rounds, nearly every time.
- Price the next one with rounds named. Two rounds included, further rounds at an hourly figure you now know is real.
That last step changes client conversations more than any rate increase. It is specific, it is fair, and it comes from a measurement rather than a mood. Turning receipts into a simple invoice line covers the mechanics of getting from a stack of days to a line on an invoice.
Photographers face a near-identical problem with a different shape of work, and Time tracking for photographers: how long editing a shoot takes is worth reading alongside this if you shoot as well as edit.
Questions
Does export time count as work? Only if you are at the machine doing something. An unattended export is machine time. Track it if you want to understand your calendar, but keep it out of the hours you divide into your fee.
How do I separate two projects edited in the same week? The app record gives you hours per day and per app. Your own one-line log assigns them to projects. There is no per-project tracking in a frontmost-app record, and pretending otherwise would produce worse numbers than a notebook.
Is one project enough to quote from? It is far better than zero and not yet a pattern. Measure three of the same format before you trust the ratio, and keep the outlier projects labeled so they do not distort the average.
What about time spent in a browser looking for music? That is edit time, and it will show as browser time. This is one of the cases where a big browser line is genuine work rather than distraction, which is worth remembering when you read the day.