A music video storyboard runs on timecode, not page numbers. You mark the song's sections, decide how many panels each one earns, then draw only the frames the edit cannot survive without. Every panel carries a time in and a time out. If you want them drawn for you, that is what the AI storyboard generator is for, and if the shots themselves will be generated, the AI music video method picks up where this board leaves off.
What changes from a film board is that your structure arrived finished. The mix already decided where the turn happens, and no amount of boarding moves it.
The track is the spine
Load the song into whatever timeline you edit in, drop a marker at every section change, and write the timecodes down before you sketch anything. That list is your board's table of contents. If you have not worked out what each section owes the viewer, the story beats version of this problem covers the structural side, and the boards should come after the treatment is signed off, not before.
Panel counts I budget against, for a three and a half minute track:
| Section | Runs | Panels | What those panels have to prove |
|---|---|---|---|
| Intro | 0:00 to 0:18 | 4 to 6 | The image system: place, palette, who we are watching |
| Verse 1 | 0:18 to 0:52 | 8 to 10 | Behavior, and the first look at the performance setup |
| Chorus 1 | 0:52 to 1:20 | 6 to 8 | The signature image, at full strength, drawn properly |
| Verse 2 | 1:20 to 1:54 | 8 to 10 | The same world escalated, never the verse 1 board reused |
| Chorus 2 | 1:54 to 2:22 | 4 to 6 | Variations only. Mark the rest "as chorus 1" |
| Bridge | 2:22 to 2:48 | 6 to 8 | The rupture. This is the section people cut first and regret |
| Final chorus | 2:48 to 3:30 | 6 to 10 | The signature image with the meaning moved |
Forty to sixty panels for a three minute video is a working board that gives the crew enough visual scaffolding without over-prescribing. Under thirty and you are asking the crew to invent on the day, filling gaps the board left open. Over eighty and you are drawing shots you will never light, wasting prep time on coverage that won't make the cut.
Panel density is a function of shot length, not ambition
Music videos cut faster than anything else you will board, demanding a higher panel-to-runtime ratio than narrative or commercial work. Divide the song's length in seconds by the average shot length you are aiming for and you get the real number of shots the edit will need, which dictates how many setups to plan.
| Average shot length | Shots in a 3:30 track | What that edit feels like |
|---|---|---|
| 1.0 s | 210 | Hard-cut performance, montage energy, exhausting past two minutes |
| 1.5 s | 140 | Standard pop-video pace |
| 2.0 s | 105 | Pop with room to breathe on the choruses |
| 3.0 s | 70 | Narrative-led, needs blocking that holds the frame |
| 4.0 s | 53 | One-idea-per-shot, closer to short film grammar |
| 6.0 s | 35 | Long-take territory. Board every single one |
You are not drawing 140 panels. You draw the shots that carry an idea and you tag the rest as coverage of an already-drawn setup. A repeated chorus angle gets one panel and a note saying it comes back at 1:54 and 2:48. That note is worth more than the drawing, because it tells the first AD the setup stays up.
What a music video panel carries that a film panel doesn't
Six fields. Skip any of them and someone on the day will guess.
| Field | Why it exists |
|---|---|
| Time in / time out | The only address a panel has. Shot numbers drift, timecode doesn't |
| Lyric or beat cue | "on the snare", "second 'I know'". The editor works to this, not to your shot number |
| Camera fps | Decides the whole day, because it decides playback |
| Playback speed | The number the playback operator sets. Wrong here means unusable sync |
| Look block | Wardrobe, hair, location. Groups the panels into shooting order later |
| Performance or concept | Which of the two passes this frame belongs to |
The lyric cue is the field people leave out and the one editors ask for first. A panel that says "wide, dolly in" is a suggestion. A panel that says "wide, dolly in, lands on the downbeat of the second chorus" is a shot.
The storyboard template generator on 熙熙.ai, captured 3 September 2026. The aspect ratio menu covers 16:9, 1.85:1, 2.39:1, 4:3, 1:1 and 9:16; panels per page run 2, 3, 4, 6, 8, 9 or 12, on Letter or A4.
Use two panels per page for the hero frames you need the record label to look at, and nine or twelve for the coverage grid the crew carries. Same board, two printouts, and nobody on set is flipping through forty pages to find the bridge.
Two passes, boarded separately
Every music video is at least two shoots wearing one name. The performance pass covers the artist singing the whole song, usually more than once, in one or two setups. The concept pass covers everything else: the narrative, the dancers, the inserts, the location work.
Board them as separate documents that share a timecode column. Then the performance pass gets a coverage plan rather than a per-shot board, because you are shooting continuous takes of the full track and choosing angles, not shots. Wide for the verses, 50mm for the choruses, long lens for the bridge, one handheld pass for safety. That is four panels covering three and a half minutes, and it is honest about how that footage gets made.
The concept pass is where the drawing goes. Those frames get boarded the way a storyboard from a screenplay gets boarded, one panel per idea, because nothing in that footage repeats.
Speed: the math that decides your shoot day
Slow motion on a lip-sync shot is not a post decision. If the artist mouths the words at normal speed and you shoot at 48 fps, the mouth is wrong forever. The fix is old and it still works: speed the playback up by the same ratio you are overcranking.
Screen speed equals timeline fps divided by capture fps. Playback speed equals capture fps divided by timeline fps. Two divisions, and they decide what the playback operator loads on the day.
| 拍摄帧率（拍摄帧率（Capture fps）） | 时间线帧率（时间线帧率（Timeline）） | 银幕播放速度（银幕播放速度（Screen speed）） | 现场回放速度（现场回放速度（Playback speed on set）） |
|---|---|---|---|
| 48 | 24 | 50% | 2x |
| 60 | 24 | 40% | 2.5x |
| 96 | 24 | 25% | 4x |
| 120 | 24 | 20% | 5x |
| 50 | 25 | 50% | 2x |
| 12 | 24 | 200% | 0.5x |
Undercranking is the row people forget. Shoot at 12 fps against a half-speed track and the performer has to mouth the words in slow motion, which looks absurd in the room and cuts beautifully.
Give the artist a pitch-corrected version of the sped-up track, because nobody can sing along to a chipmunk, and give the playback operator the raw sped-up file so the timing stays exact. Write the conform on the panel too. In post the clip gets reinterpreted at the timeline rate rather than ramped by eye, and Adobe's own Interpret Footage documentation covers what that does to duration. Frame counts for a specific ramp are quicker on the video frames calculator than in your head at 6am.
Try it free
Try 熙熙 for free on your script
It is free. Import your existing project, get a clearer view of your outline, and regain control of your story structure in minutes.
Start FreeReorder the boards before you print them
The board is written in song order. The day is shot in setup order. Those are two different documents and the crew needs the second one.
Sort your panels by look block, then by location, then by lighting state, and print that version for the floor. Keep the song-order version for the director and the editor. On a one-day video with two locations and three wardrobe changes, the shooting-order board will jump from 2:48 to 0:14 to 1:31 in the first hour, and that is correct. The timecode field is what lets you do this without losing anything.
The order also tells you what to protect. Whatever you shoot last is what gets cut when you run out of light, so the bridge should not be last.
Mark the vertical crop while you are still drawing
A 9:16 deliverable now appears in nearly every project brief, yet it consistently remains an afterthought in the planning phase. If your deliverables list includes one, establish a clearly marked vertical safe area on every storyboard panel where the primary subject is not centered in frame, and candidly assess which specific shots will be lost or compromised in the crop. A two-shot positioned at the outer edges of a 2.39:1 frame simply cannot survive the transition to vertical format intact. It is far more valuable to identify these incompatibilities during the drawing stage than to discover them three weeks downstream during the social media edit when options for recourse are limited.
The same panel grid structure in 9:16 format requires approximately ten minutes to complete for the twenty frames that genuinely carry weight in the final vertical cut and determine whether the story holds together.
Checklist before the boards leave your hands
- Every panel has a time in and a time out
- Every panel names a lyric or a beat, not just a bar number
- Capture fps and playback speed written on every high-speed or ramped panel
- Repeated chorus setups tagged with the timecodes they come back at, not redrawn
- Performance pass and concept pass separated, sharing one timecode column
- A shooting-order print exists alongside the song-order print
- Look blocks labeled so wardrobe can group the day
- Vertical safe area marked on any panel that will be cropped
- Total shot count sanity-checked against your target average shot length
- One page that lists only the frames the video cannot be delivered without
That final line of identification is the one that ultimately saves the production when circumstances deteriorate on set. When the shooting day inevitably collapses under time pressure or unforeseen complications, and it will collapse, someone on the crew must possess definitive knowledge of which specific frames are absolutely non-negotiable for the edit to function.
Tools for this
The storyboard template generator tool outputs panel grids formatted in the precise aspect ratio you are capturing, including both 9:16 vertical and 2.39:1 widescreen formats. If your workflow involves generating frames through AI tools rather than hand-drawing them, the storyboard-first workflow for AI video methodology is the structured process that maintains the boards' utility and reference value once the generated images are produced, while generating a board from a script addresses the narrative design side specifically for early concept passes that incorporate dialogue and story beats.
FAQ
How many panels does a music video storyboard need?
Forty to sixty total panels for a three minute video piece, systematically covering the shots that carry a distinct narrative idea or visual beat forward. The final edited sequence will inevitably contain far more cuts than that panel count, because repeated performance angles from the same camera position and additional coverage of an already established setup do not require their own dedicated panel in the board. Tag these repeated setups with the specific timecodes where they return in the sequence instead of redrawing them as redundant panels.
Do I storyboard the performance footage?
Performance coverage doesn't follow a shot-by-shot breakdown. Instead, the artist performs the entire song in continuous takes, and the director plans coverage around which lens focal length, camera angle, and which musical section to emphasize. Typically four to six carefully chosen panels provide sufficient coverage for a complete performance pass. The concept pass—scenes with narrative or stylized visual ideas—is where traditional per-shot storyboarding proves its value and justifies the detailed planning time.
What playback speed do I use for slow motion lip sync?
To calculate your slow-motion playback speed, divide your capture frame rate by your timeline frame rate. For example, shooting at 48 fps for a 24 fps timeline yields 2x slow motion, 96 fps produces 4x, and 120 fps delivers 5x playback speed. Provide the performing artist with a pitch-corrected reference track so they can lip-sync and perform in sync, while the playback operator works with the raw sped-up audio file that matches the high frame rate capture.
Should the storyboard follow the song order or the shooting order?
Print both versions separately—they serve different departments. The song-order storyboard, arranged by the track's musical progression, is the reference document for the director and the editor during post-production. The shooting-order board, reorganized by look block, lighting setup, and physical location, becomes the working document for the production floor crew. Timecode notation on every panel is the linking mechanism that allows seamless cross-referencing between both versions without losing any shot or causing confusion.
What goes on a music video panel that a film panel doesn't have?
Time in and time out, the lyric or beat cue the shot lands on, the capture frame rate, and the playback speed—these are the coordinates of a music video panel. Film panels address shots by scene and shot number, a hierarchical system inherited from the script. A music video panel is addressed by timecode, because that is the only reference the song, the board, and the edit all agree on, and the only one that survives from pre-production through final color.
Can I storyboard a video before the track is finished?
You can board the treatment, not the video itself. Section lengths move during the mix—a bridge stretches, a chorus gets trimmed—and a board built on provisional timecodes has to be rebuilt anyway. Wait for the version that will actually be released, the locked mix with final BPM and arrangement, then mark the sections and start. Boarding earlier means boarding twice.
Final Step
Build your next script with 熙熙
Move from ideas to production-ready pages faster with timeline-native writing and AI-assisted story flow.
Try 熙熙
