The question almost always arrives at the end of the first meeting, and almost always with a date already decided behind it. «We need this for the March event.» The honest answer isn't a number: it's a calendar with phases, because the time a video takes isn't spread evenly, and because some phases shrink well and others don't shrink at all.
What follows are typical durations for a mid-scale brand video — one or two shoot days, one or two locations, a handful of people on camera, a main piece and a few short versions. These aren't rules. They're the starting point for working out where your own schedule is lying to you.
Phase 1 — Brief and concept: one to two weeks
This phase produces nothing you can see, which is why it's the first to get cut. Its job is to turn a vague request into a concrete decision: what does this video have to do, for whom, where will it live, and what idea holds it up.
The real work here is reduction. A brand almost always arrives with five things to say and leaves with one. Getting to that one takes conversation, disagreement and, sometimes, going back on yourself. A week is enough if there's already internal clarity. Two is more realistic if the decision involves more than one person.
What drags this phase out is never creative difficulty. It's the diary of whoever decides. If the first version of the concept sits waiting for a meeting that only exists ten days from now, you've lost ten days in the place where they're cheapest to avoid.
Phase 2 — Script and pre-production: two to three weeks
This is where the quality of the video is decided, and it's the most misunderstood phase of the schedule. From the outside, it looks like nothing is happening. In practice, nearly everything is.
The script takes its share: writing it, reading it aloud against a stopwatch, cutting, approving. But the script is only the beginning. Out of it come the shot list, the number of hours needed, the locations, the people who speak, the kit that has to be in the right place. Each of those generates tasks with deadlines of their own, and many of them don't depend on the production company at all.
- Location recces. They have to happen at the hour you'll be filming, which means you can't do them all in one afternoon.
- Permissions. Building management, shopping centre administration, council permits for public streets. None of these bodies answer the same day, and some take weeks.
- People's availability. Getting three directors into the same morning is frequently the slowest item in the whole production.
- Casting, hair and makeup, wardrobe, if there are people on camera who don't work for the company.
Two weeks is the comfortable minimum when the locations are your own and the people are internal. Three weeks is normal the moment a permit application or an outside person enters the picture.
Phase 3 — The shoot: one to three days, but not in just any week
The most visible phase is the shortest. A typical brand video is resolved in one or two days. More than that only when there are several distant locations, a large number of people on camera, or scenes that depend on specific conditions.
What matters for the schedule isn't the length of the shoot — it's its rigidity. A shoot day is a fixed date with ten paid people, hired equipment and booked locations. If it slips, it doesn't slip by a day: it slips to the next date on which all those pieces line up again, and that can be a fortnight later.
Which is why the shoot is the one point in the calendar worth treating as fixed and building everything else around, backwards and forwards.
Phase 4 — Edit and first cut: one to two weeks
After the shoot there's always an interval nobody counts: copying, organising and watching all the material. On a day of interviews, that's hours of viewing before the first shot gets cut.
The edit itself starts with a structure — the order of the ideas, the sentences that survive, what opens and what closes. That first version has no colour treatment, no sound mix and no final graphics, and that's deliberate: it exists so you can discuss whether the story is right, not whether the blue is pretty. Polishing before the structure is approved is work you throw away.
A week is enough for short, well-structured material. Two is normal when there's a lot of raw footage, or when the structure still has to be discovered in the edit.
Phase 5 — Revision rounds: two to three weeks
This is the phase where schedules die, and almost never through the fault of whoever is editing.
Two or three rounds of changes is reasonable. Each round has two halves: the time of the person making the changes and the time of the people watching and responding. The second half is usually longer than the first. A change that takes a day to execute then sits for three days waiting for somebody to confirm it's fine.
What turns two weeks into two months is always the same thing: somebody appearing halfway through who wasn't in the original conversation. It isn't bad faith — it's that anyone seeing a video for the first time at the third round will naturally question decisions taken back at the concept stage. And undoing a concept decision at that point means going back into the edit, sometimes back to the shoot.
Deciding, before anything starts, who has the final word is the single measure that saves the most time on a video project. More than any tool, any software and any amount of goodwill.
Phase 6 — Finishing and deliverables: one week
With the structure locked, the finishing work begins: colour correction and grade, sound mix, graphics and captions, licensed music, subtitles, and then the export of each version.
The part usually forgotten in the time budget is the versions. The main piece, the short cut for social, the vertical version, the version with burnt-in subtitles, the clean version with no text for another language. Each one is a different edit, not a different export. A video that becomes six versions adds days to the final phase — and if it's only requested at the end, it adds them after the deadline has already passed.
The whole schedule, added up
Adding the phases together, a mid-scale brand video typically takes eight to twelve weeks from first meeting to final delivery. Projects with weather-dependent exteriors, several cities or complex licensing stretch beyond that easily.
The distribution is worth looking at: the shoot, which is the part everybody pictures when they think about video, takes one or two days of that total. Everything else is preparation and desk work.
What you can speed up
Not everything is untouchable. There are four legitimate ways to shorten a schedule, and none of them involves asking the team to work faster.
- Reduce scope. One location instead of three. Two people on camera instead of six. One piece instead of five versions. It's the only method that cuts time without cutting quality.
- Decide faster. Book the approval meetings before there's anything to approve. One day of response instead of five repeats itself in every phase and adds up to weeks.
- Handle the admin early. Permits, image releases, people's diaries. None of this needs a locked script and all of it can start in week one.
- Overlap phases carefully. While the script is being approved, the location recces can be happening. Not everything is sequential.
What you can't speed up
And then there's what doesn't give, however hard you push.
Capture time. A good interview doesn't happen in twenty minutes, because the person needs ten of them to stop being nervous. A well-lit shot takes as long as the lighting takes to build. The alternative isn't doing the same thing faster — it's doing less.
Watching the material. To choose the best sentences, somebody has to listen to all of them. There's no shortcut and no way to delegate it to someone who doesn't know the project.
The distance between versions. An edit watched immediately after it's made isn't watched properly. Reviewing it the next day, with your head out of it, is the difference between noticing the opening is slow and not noticing.
The sum of two things that have to happen one after the other. The sound mix happens with the edit locked. The subtitles happen with the final cut. There's no parallel working at the end of a project, which is why the last week doesn't compress.
The good date is the one chosen early
The most expensive scheduling mistake in video isn't giving production too little time. It's giving the internal decision too little time and then demanding that production make up the delay in the phase where making it up is impossible.
If you have a launch date, count backwards from it, with the finishing untouchable at the end and the shoot fixed in the middle. What's left over is your thinking time — and if nothing is left over, you already know it isn't production that will have to be faster: it's the scope that will have to be smaller.