Write a shot list in six columns and then reorder it. The list is written in story order so it makes sense, and shot in setup order so it is achievable, and lists that never get reordered are why days overrun. Every row should be specific enough that someone else could capture it without asking you a question.
The six columns
| Column | What goes in it |
|---|---|
| Number | Stable reference, never renumbered mid-project |
| Framing | Wide, medium, close, extreme close, over-the-shoulder |
| Subject and action | What is in frame and what happens |
| Camera | Static, push, pan, handheld, or explicitly "any" |
| Duration | Seconds, realistic rather than aspirational |
| Note | The link back to the brief or the moodboard |
Two optional columns earn their place on bigger jobs: setup, so you can sort by it, and status, so the day is trackable.
Writing the rows
Be specific about the action, loose about the poetry. "Hands twist the lid, thread visible" is a row. "Capture the satisfaction of opening" is a mood, and moods do not get captured, they get produced by specific actions.
Give every row a duration, even a guess. The sum is the runtime, and a list that sums to 90 seconds for a 30-second piece is a plan to shoot 60 seconds of material you will not use.
Mark the hero rows. The two or three shots the piece cannot exist without. When the day runs short, the marked rows are the ones you protect.
Add coverage rows deliberately. Two or three per setup: cutaways, textures, reactions, hands. They cost minutes on the day and save the edit. See what is b-roll.
Reorder before the day
Sort by setup, then by talent, then by anything requiring a change. Every lighting change costs 30 to 60 minutes and every wardrobe change costs about the same, so a list shot in story order pays that cost repeatedly for no benefit.
Then check the reordered list against the story order once more, to catch the continuity problem that reordering just created.
Mark what does not need capturing
The column that matters most in 2026. Go through the list and mark rows as capture or generate. The reliable split, and it holds across projects: capture the people, the products, and any mechanism a viewer needs to believe; generate environments, establishing shots, atmosphere, seasonal variants, and context that surrounds the real material.
Marked rows tell you how long the day actually is. A twenty-row list where seven rows are generatable is a substantially shorter day, and the seven get produced afterward from a reference frame so they match what you shot.
Canon 8frame prices, a credit at $0.01 in a top-up pack: a generated still is 6 to 21 credits, an 8-second Veo 3.1 Lite clip with audio is 54, a 5-second Kling v3 Standard clip with audio is 86. Seven generated rows land between 40 and 600 credits depending on stills or motion, against an hour or more of shoot time each.
An example row set
For a 20-second product piece:
- Wide, product on surface, static, 3s, hero row
- ECU, texture of the surface, static, 2s, generatable
- Medium, hands enter and lift product, handheld, 4s, hero row
- ECU, mechanism opening, static, 4s, hero row, must be captured
- Wide, product in use in room, any, 4s, generatable
- Medium, result held to camera, static, 3s
- Coverage: three cutaways, any framing, generatable
Four capture rows and three generatable ones, which is a two-hour shoot rather than a day.
FAQ
How detailed should a shot list be? Detailed enough that someone else could shoot it. If a row needs you present to interpret it, it is not finished.
Should a shot list include the storyboard frames? On projects where the framing matters, yes, side by side. On documentary-style shoots, the list alone is better since the board would be fiction.
How many coverage shots per setup? Two or three. They are the cheapest insurance in production.
Lists work best beside the frames they describe and the generations that fill them in. The 8frame canvas is free and unlimited, and generation is paid from $19/month. See also what is a shot list.