Correction, 2026-10-02: the 2026-10-01 note below said 8frame has no timeline. That was wrong: the canvas has none, but 8frame's separate Video Editor mode has a timeline with video, text, caption and sound layers, and we have fixed the note. The old call to action was wrong only in sending you to the canvas to open a timeline. This pass also removed or softened lines we cannot source or have not measured: that agencies and studios have relied on animatics "for decades", that the length of a voiceover line is the "truest" guide to a beat, and that testing at the animatic stage means production money "only" goes to ideas that proved they work. The two examples now read as worked illustrations, not reports of real sessions, and the FAQ adds that Veo animates a start image at 8 seconds only, in 16:9 or 9:16. The FAQ now also says that the Montage node joins the picture only: the merged file has no sound, so temp voiceover and music go on the Video Editor timeline or in your own editor.
Correction, 2026-10-01: an earlier version priced its two examples at "roughly $9" and "under $4" with times of "an afternoon" and "90 minutes"; those were not measured runs. It also said an outsourced animatic used to mean a four-figure fee and several days, which we cannot source, and its call to action sent you to the canvas to "open a timeline"; the timeline is in 8frame's separate Video Editor mode, not on the canvas. The costs below are computed from our price list.
An animatic is a storyboard put in motion and cut to its real timing, with each panel held for its true duration and temp voiceover and music laid underneath, so a team can feel a commercial's pacing before it is produced.
It sits between the still storyboard and the finished spot. The board tells you what happens and in what order. The animatic tells you whether the cut actually works: whether the joke lands on time, whether the product reveal has room to breathe, whether the whole thing runs long. It is the last cheap checkpoint before a production budget is committed, and AI has made its motion step cheaper.
How an animatic works
An animatic starts from an approved storyboard. Each panel becomes a timed segment: held as a still, given a simple move, or animated, then trimmed to the duration it will occupy in the final spot. Those segments are cut together in sequence so the piece plays at its true length, such as 15, 30, or 60 seconds for a commercial.
Two audio layers turn the moving panels into a testable spot. First, temp voiceover, a scratch read of the script whose job is to lock timing rather than deliver a final performance. The length of each line is a good early guide to how long a beat needs. Second, a music bed that sets the tempo and mood, because pace and emotion depend heavily on the track underneath. With both layers in place, the animatic shows pacing problems that are hard to see in a static board.
Traditionally an animatic meant limited motion: panels panned and zoomed, held on cuts, with the occasional simple animation. An AI version can use image-to-video generation to give each panel real, motivated movement, so the animatic reads closer to the finished cut while still being clearly a rough, testable draft rather than the final spot.
When you use an animatic
You use an animatic after the storyboard is approved and before the shoot is booked. Its purpose is to de-risk the expensive part of production by proving the cut works while changes are still cheap.
The clearest case is pacing. A 30-second spot depends on its timing, and timing is hard to judge from stills. The animatic is where you discover that panel six holds two seconds too long or the run time is four seconds over, and fix it by moving panels instead of reshooting.
The other main case is testing. Because an animatic is far cheaper than a finished spot, teams put it in front of an audience, a real research panel or a synthetic one, to check whether people follow the story, whether the brand registers, and whether attention holds to the end card. Testing at the animatic stage means more of the production budget goes to ideas that have already been tested. That is the economic argument for the format.
AI lowers the cost of the motion step. The same tools that generate the board can animate it, so animating every panel costs credits rather than a separate vendor's time. The format did not change; what changed is how cheaply a team can try it before the crew is booked.
Examples
Both examples are worked illustrations with costs from 8frame's price list, not reports of real sessions.
30-second beverage spot animatic. An eight-panel board for a coffee brand is animated one panel at a time with an image-to-video model, each clip trimmed to its scripted length. On 8frame, eight 5-second Wan 3.0 clips at 480p are 8 x 24 = 192 credits; on Kling v3 Standard with sound off they are 8 x 57 = 456. A temp voiceover of the script is recorded on a phone and laid across the timeline in an editor, such as 8frame's Video Editor mode, and a Lyria 2 music bed adds 13 credits per generation (a 30-second clip). That puts the generation cost at roughly 205 to 469 credits, about $2 to $5 at pack rate, before editor time. Say the team watches the full 30 seconds and finds that the espresso-pour beat rushes into the product reveal: they add a one-second hold, re-cut, and send the re-timed cut to a test panel before the shoot is scheduled.
15-second social spot animatic. A four-panel board for a vertical 9:16 app-install ad is animated with subtle motion per panel (four 5-second Wan 3.0 clips at 480p: 96 credits), temp VO is added as a text-to-speech read to test the script's clarity at speed (ElevenLabs on 8frame is 13 credits per started 1,000 characters, and a 15-second script is well under that), and a driving music bed is set for a fast cut (Lyria 2, 13 credits). Total: about 122 credits. If watching it back shows that the call-to-action end card flashes by too fast to read, extend the last panel by half a second and tighten the preceding cut to keep the total at 15 seconds. That is a comprehension problem that would be expensive to fix after production.
FAQ
Can I cut the animatic inside 8frame? Yes. On the canvas, the Montage node joins 2 to 10 clips with transitions for 10 credits a run, which is enough for a rough look at the picture: it uses at most 15 seconds of each clip, and the merged file has no sound. For timing with temp voiceover and music, 8frame's separate Video Editor mode has a timeline with video, text, caption and sound layers; any desktop editor works too.
What does the motion layer cost? It depends on the model and length: a 5-second clip is 24 credits on Wan 3.0 at 480p and 57 on Kling v3 Standard without sound, and an 8-second clip is 168 on Veo 3.1 Fast with sound. Veo animates a start image at 8 seconds only, in 16:9 or 9:16.
Related concepts
The AI storyboard workflow is the step before the animatic: breaking a script into shot beats and generating consistent frame stills you arrange into a board. The board's shot durations and voiceover cues become the timing map the animatic is built on.
The AI animatics workflow is the full production recipe for making one: animating each panel, adding temp voiceover and music, and cutting to timing.
Sources
- Wan 3.0, Kling v3, Veo 3.1, Lyria 2, ElevenLabs TTS and Montage prices (calculateWan3Cost, calculateKlingV3Cost, calculateVeo31Cost; Lyria 2 13 per generation; TTS 13 per started 1,000 characters; Merge Videos 10 per run); credit value at pack rate: 8frame tool registry and billing catalog, read 2026-10-02
- Veo with a start image (8 seconds only; 16:9 and 9:16) and the Video Editor mode (a timeline with video, text, caption and sound layers): 8frame front-end and video generation code, read 2026-10-02
Turn your next storyboard into a tested animatic: the image-to-video models and workflow templates live on one 8frame canvas. Open the canvas, animate your panels, and feel your spot's pacing before you book the shoot.
Try it without signing up: the 8frame canvas opens on a real board. The canvas is free and unlimited; generation is paid from $19/month.