Video Team Capacity Planning: How to Know Who's Actually Free

A client asks whether you can take on three more videos this month. You look at the board. It is full, but boards are always full. You say yes, because you usually can, and because saying no to a good client is expensive.

Three weeks later your best editor is working the weekend, two deliverables slip, and you discover you were never going to make it - the maths said so on the day you agreed. You just did not have the maths in front of you.

That is a capacity problem, and it is different from a project management problem. Your board already tells you what is due and who owns it. Capacity planning tells you whether the hours exist at all.

Why a Task Count Is Not Capacity

Most teams estimate capacity by counting tasks. “Maya has six tasks, Tom has four, so give it to Tom.” The problem is obvious the moment you name the tasks: Tom's four are feature-length grades, and Maya's six are fifteen-second cutdowns.

Task count is a proxy that fails exactly when you need it most - during a crunch, when the mix of work is at its most uneven. The only unit that survives that is the hour.

Real capacity planning needs three inputs, and most teams are missing at least two:

1. A capacity per person, in hours per day.Not “full-time.” A number. Six productive hours is a more honest figure than eight for most editors, once you subtract calls, revisions triage, and context switching.
2. An estimate per task, in hours.Rough is fine. “About a day” beats nothing, and it gets more accurate the moment you start comparing estimates against tracked time.
3. A window per task. A start date and a deadline, so the hours can be spread across the days that actually exist between them.

With those three, the answer to “can we take this on?” stops being a feeling.

Spread the Hours, Don't Stack Them

Here is the single most common modelling mistake: putting a task's full duration on its deadline.

A twelve-hour edit due Friday is not twelve hours of work on Friday. It is a twelve-hour block that has to land somewhere between now and Friday. If your plan shows it as a Friday item, every Friday in your schedule looks catastrophic and every Wednesday looks empty - so you stop trusting the plan, which is the same as not having one.

Spread the estimate across the working days in the task's window instead. Twelve hours between Wednesday and Friday reads as four hours a day, which you can compare against a real daily capacity. Now overbooking is visible as a number: eleven hours booked against an eight-hour day is three hours you do not have.

Count the Things That Aren't Tasks

Editors do not spend their week only on deliverables. If your capacity model ignores everything that is not a task, it will be optimistic by 20–40% and you will never work out why.

The usual missing hours:

• Time off. Not a reduced day - a zero. A person on vacation has no capacity, and any plan that quietly books them is lying.
• Shoot days. An editor on set is not editing. Shoots need to consume capacity the same way a task does.
• Admin, calls, and client management. If a senior editor spends a day a week on calls, their capacity is four days, not five. Book it as a recurring block rather than pretending.
• Revision rounds you haven't received yet. If your average deliverable takes 2.4 rounds, the second round is not a surprise - it is a statistically certain future demand on the same person. Leave room for it.

On that last point: the cheapest capacity you will ever find is the revision round you did not have to do. Most of our common agency mistakes end up expressed as unplanned rework, and rework eats capacity that was already promised elsewhere.

The Warning Has to Arrive Before the Promise

A capacity plan that you have to go and read is a capacity plan you will read after the fact. The value is entirely in the timing: the warning is useful at the moment you are about to say yes, and worthless in the retrospective.

Practically, that means the check has to fire where the commitment is made - when you set a deadline, when you change an estimate, when you assign work. Something has to say, in the moment: Tom is over capacity on Tuesday - eleven hours booked against eight.

And it should be a warning, not a wall. Crunch weeks are a legitimate business decision; agencies take them deliberately all the time. A planning tool that refuses to let you overbook is a tool your team will route around within a fortnight. The job is to make the cost visible, not to make the choice for you.

Estimates Get Accurate by Being Compared

“We can't estimate” is the most common objection to capacity planning, and it is almost always a chicken-and-egg problem. Estimates are bad because nobody compares them to anything; nobody compares them because they are bad.

Break the loop cheaply. Estimate roughly, track time on the task, and look at the gap monthly - not per task, per type of task. You do not need to know how long this reel takes. You need to know that reels average 3.5 hours and you have been budgeting two.

Within two months you will have per-format numbers that are good enough to plan with, and - usefully - good enough to price with. See our breakdown of how long editing actually takes per video type for realistic starting figures while you build your own.

Capacity Is a Pricing Input, Not Just a Scheduling One

Once hours are real, they stop being only a scheduling concern. An editor's hours have a cost, and different roles have different costs - a colorist and a rough-cut editor are not interchangeable line items on the same project.

The teams that run this well end up with a chain that closes: estimate → booked hours → tracked hours → cost → margin. When that chain is intact, “we're busy” and “we're profitable” stop being the same sentence, and you find out which clients are consuming capacity they are not paying for. That is usually the finding that changes the business. Our guide on paying video editors covers the rate side of that equation.

A Practical Starting Point

You do not need a six-week rollout. A workable version of this takes an afternoon:

1. Set an honest daily capacity for each editor. Start at six hours, not eight, and adjust once you have tracked data.
2. Put a rough hour estimate on every open task. Anything is better than nothing; you will fix them with real numbers later.
3. Give each task a start date as well as a deadline, so the hours can spread.
4. Block the non-task time - vacations first, then shoots, then recurring admin.
5. Look at the week. The overbooked days will be immediately obvious, and they will not be the days you expected.

Timeliner does this natively in its Workload view: each member carries a daily capacity, task estimates spread across the working days between start date and deadline, unscheduled work sits in a backlog rail you drag onto a person's day, and time off and production blocks consume capacity like anything else. When a change overbooks someone, the warning names them, the day, and the size of the overage - before you commit. Booked work can also be subscribed into Google Calendar, Apple Calendar, or Outlook, so editors see their week where they already look.

Pair it with workflow automations and the plan starts maintaining itself: a task entering a stage can set its own estimate and start date, so work books itself instead of waiting for someone to remember.