Motion and video

What a storyboard is, and how to draw one that is actually useful

A storyboard is a video's shot list, where each shot is pinned down with a very rough sketch and a few lines of notes: what is in frame, what the camera does, what is said, and how long it lasts. Its value is not in the quality of the drawing but in pulling disagreements out before the camera is switched on.

  • Lesson 6 of 10
  • Beginner
  • Free, no signup

A short promo, broken into sequences and shots

  1. Hook

    2
    • Close, static

      One detail of the product, unexplained. Two seconds.

    • Medium, slow pan

      The first line of voiceover lands on this shot.

  2. Problem

    2
    • Wide, static

      The scene where the problem is visible, no dialogue.

    • On-screen text

      Motion graphics. Three words, not a sentence.

  3. Solution

    2
    • Close on hands

      The product in use. Cut on action here.

    • Medium, face

      The reaction. The longest shot in the piece, three seconds.

  4. Call to action

    1
    • Logo and address

      Motion graphics, no sound. Silence here is a decision.

This division is one common shape for a product promo, not a rule. For documentary, tutorial and interview video, the sequences are something else entirely.

Last checked: Facts and tool names in this lesson are re-checked against their sources on this date.

What is a storyboard, and what is it not?

A storyboard is a shot-by-shot map of a video: each shot gets a cell, the cell holds a very rough sketch of the frame, and beside it you write what happens.

What it is not: artwork. This is where people usually stall, because they assume that not being able to draw means they cannot board. Frames drawn with four rectangles and two stick figures do exactly the same job, provided the notes beside them are precise.

What it is: a decision document. The board is where expensive decisions get made before they become expensive. Whether this shot is tight or wide, whether the camera moves, whether the line is on-screen text or voiceover: on paper each of those costs a few minutes, and on set or mid-edit each can cost half a day.

And a distinction many people blur: a shot list and a storyboard are not the same thing. A shot list is a text table and it is enough to plan a shoot day. A board is that table plus a picture, and what the picture does is expose misunderstanding. A client and an editor can read the same sentence and see two different things; the frame gives that away on the spot.

Six rough paper storyboard frames with arrows drawn in light, lit red, green and blue

Why is a badly drawn board better than no board?

Because the cost of a change multiplies at every stage, and the board is the cheapest stage. Changing a frame on paper takes minutes. Changing that shot after the shoot means going back to the location. Changing it after the animation means throwing away work that has been built.

We are not writing that with percentages, because the real numbers depend on the project, but the order never changes, and anyone who has once had to reshoot a scene does not need convincing.

There is a second benefit that gets said less and matters more in client work: a board makes sign-off possible. Once the client has seen and approved the list of frames, "this is not what I wanted" turns into a specific conversation about frame seven. Without a board, that same sentence means the whole project again.

And a third benefit, this one for the editor: a board tells you what you failed to shoot. With the frame list in front of you, the end of the shoot day is when you notice the connecting shot between scenes two and three was never taken. That short, unimportant-looking shot is the one that stops everything at the edit desk.

What does each frame have to note?

Five things, and none of them is optional.

Frame size: wide, medium or close. That single word decides where the viewer looks, and most on-set arguments start from its absence. Camera movement: static, pan, tilt, dolly or zoom. If it is static, write static; leaving it blank means everyone guesses. What is said or written: the exact line, not a summary of it. Approximate length: in seconds. The sum of this column is what tells you your thirty-second video is actually fifty seconds, and it is better to learn that here than at the edit. Sound: music, effect, silence. Silence is a decision too and it gets written down.

In the board drawn at the top of this lesson, the seven shots of a short promo run two to three seconds each, and it is the sum of that column that tells you how long the video actually comes out.

There is a sixth column that only means something in client work, and we always add it: how this shot gets made. Filmed, animated, archive footage, or screen recording. That one column is what stops a board that is perfect on paper and twice the project's budget.

What you must not write in a frame matters as well: a description of a feeling. "The scene should be warm and intimate" is a sentence that everyone executes differently. If warm and intimate matters, translate it into something executable: side light, tight frame, wide lens, no music.

What we write beside each frame, and what we do not

Frame sheet: what gets writtenboard

Goes on the frame

  • Frame size: wide, medium or close
  • Camera movement, even when it is static
  • The exact line that is spoken or shown
  • Approximate length in seconds
  • Sound: music, effect or silence
  • How it gets made: filmed, animated, archive

Does not go on the frame

  • Mood adjectives like warm, intimate, modern
  • Story explanation that is not visible in the frames themselves
  • The name of a program or plugin
  • Details that will be decided on set anyway

The second column is not wrong, it is just not on the board. Feeling and visual reference belong in a separate brief document.

How does a motion graphics board differ from a film board?

Both do the same job and ask two different questions. A film board asks where the camera is and who is in frame. A motion board asks what enters the frame, from which side, and what leaves at that same moment.

Which means that in motion, movement is the content of the frame, not an adjective attached to it. A frame that only shows the final composition is close to useless, because that composition may only be on screen for half a second. So motion boards usually give every important change its own frame: the first state, the middle state, the last state.

There is another practical difference: in motion, boarding and design sit closer together. Colour, type and layout, which on a film board are only hints, are on a motion board more or less the thing being made, because there is no footage coming later to change the look. That is why motion boards often follow an intermediate step called a style frame: one or two fully coloured frames that get the visual language approved, after which the rest of the frames are drawn plainly and fast.

If you are new and the project mixes both, the simplest rule is this: take the "how it gets made" column seriously, and for every animated shot draw three frames instead of one.

Film board and motion board: two different questions

Film board: where is the camera

  • One frame per shot, because the framing holds to the end of it
  • Angle, lens and camera position are the main decisions
  • Colour and light are only hints; they get made on set
  • The board turns into a location and equipment list

Motion board: what enters the frame

  • A frame per important change: first, middle, last
  • Which way elements enter and leave is the main decision
  • Colour and type are more or less the thing being made
  • Usually one coloured style frame is approved ahead of the rest

Neither is harder or more professional than the other. A mixed project needs both, and the "how it gets made" column decides which each shot is.

What do you draw the board with?

With anything that is fast and lets you move things around. Those two conditions are the whole criterion.

Paper and pencil is the fastest option and is always enough for a personal board. Its weakness is that it is not sendable to a client, and reordering frames means redrawing them. A slide or presentation tool turns each frame into a slide and solves that reordering with drag and drop; for most promotional projects that is enough. Interface design tools like Figma earn their place when the board is going to go through several versions with several people commenting, because they have comments and version history.

Our position: start with whatever you will actually open right now. A board drawn on paper in twenty minutes and sitting with the client tonight is infinitely better than the board that was going to be made in the proper tool and never was.

One warning about ready-made templates: a storyboard template usually has six cells on a page, and that alone can push you into thinking in six shots. The number of frames is decided by the story, not by the dimensions of the paper.

The fast path, with AI

The fast path for boarding has two stages, and the first one is where the time goes. A strong model turns the script or brief into a frame list with those same six columns, you correct the table, and only after the table is approved do you go near a picture. The mistake most people make is generating images first and discovering afterwards that the story was wrong, which is exactly the mistake the board exists to prevent.

  1. Turn the script or brief into a frame table with the recipe below. Use a strong model, not a cheap fast one, because this step is a narrative decision and not mechanical work.
  2. Add up the length column. If it is over the target, drop frames now rather than later; this is the only stage where dropping is free.
  3. Check the "how it gets made" column against the budget. If half the shots are animation and the project has no animation budget, the table changes here, not three weeks later.
  4. Now the picture. For a personal board, stick figures and rectangles are enough. For a board going to a client, each frame can be turned into a rough visual with an image model, subject to the two rules in the caveat below.

Copy-ready recipe

Role: director of the edit. Build a frame list from the text below.

Goal of the video: {one sentence}
Target length: {e.g. 30 seconds}
Audience: {who they are and where they see this}
Means available: {e.g. phone camera only, no actors}

Script or brief:
{paste here}

Build a table, one row per frame, with these columns:
1. Frame number and the sequence it belongs to
2. Frame size: wide / medium / close
3. Camera movement, and if it is static write static
4. The exact line spoken or shown on screen
5. Length in seconds
6. Sound: music / effect / silence
7. How it gets made: filmed / animated / archive / screen recording
8. One line describing the frame, only what is visible

Rules:
- Under the table, write the sum of the length column and how far it is
  from the target length.
- Never propose a shot that is impossible with the means listed above. If
  a shot is needed and is not possible, put it in a separate list headed
  "outside the means".
- Write no mood adjectives in column eight. Warm, modern and premium are
  banned; only what the camera sees.
- Take the lines from my text. If a line is needed and is not in the text,
  put it in a separate list headed "suggested copy" for me to approve.
- Do not produce more than {number} frames.

Before you trust the output: Two rules when you generate the frames with an image model. First, a generated frame has to show something that is actually achievable; a beautiful image of a shot this project cannot afford to film creates exactly the expectation the board was supposed to prevent, and the client will show you that image later. Second, a generated frame is an internal reference, not a deliverable: output from Google's image models carries the SynthID digital watermark, which Google itself says it can detect, so passing that board off as hand-drawn work is not an honest claim. Read the table yourself too: models are optimistic about lengths and generous in column seven.

AI in this kind of work

On storyboards, AI does two different jobs and they are not worth the same. The first is turning text into structure: the frame list, the columns, the length total. That genuinely buys time. The second is generating the frame image itself, which looks more impressive and is worth less, because the stick-figure sketch was already doing that job. Our position: delegate the first stage, and delegate the second only when the board is going to someone who will not trust a rough sketch.

Tools that actually help

  • Claude A good fit for building the frame table from a script, because it keeps the script's own lines verbatim in column four instead of rewriting them. Iran is not on Anthropic's supported-country list; the payment layer is on this site's buying-AI page.
  • Gemini When the brief is long and scattered and has to be gathered out of several documents, it is the cheaper choice. Iran is not on Google's supported-country list.
  • RGB AI image ranking Not a tool, and here on purpose: if you are going to generate the frames as images, read our image ranking table, with its criteria, sources and Iran-access column, instead of asking which model is best.

Where it backfires

The risk specific to this topic is a board that looks more expensive than the project. A frame generated by an image model has cinematic lighting and a location that were never in the budget and cannot be shot with the available kit; the client approves it and then holds the real output up against it. This is not the only way a board lies by accident, but it is the most common. The second risk is ownership and disclosure: the brief for an unreleased campaign carries the product name and the launch date, and sending it to an outside service is a decision, not a step. The third risk is technical and simple: images from Google's models carry the SynthID mark and Google itself says it can be detected, so if anyone has claimed the board was drawn by hand, that claim is refutable. None of the three is a reason not to use these tools; they are a reason to treat a generated frame as an internal reference rather than a deliverable.

Sources: Google DeepMind: SynthID Anthropic: is my data used for model training Google: Gemini Apps Privacy Hub

Where this advice stops

Not every video needs a storyboard. An interview where you do not know in advance who says what, an event that happens once, and a raw social clip you shoot and publish the same day are all slowed down by a board rather than sped up. The practical boundary: wherever more than one person has to know what is being made, or wherever reshooting is expensive, a board pays. And one more limit worth stating: a board is not a substitute for a brief. If the goal of the video and its audience are unclear, a frame list only tidies the confusion up.

From our own work

These lessons you are reading are themselves storyboarded, just with infographics instead of film. Every figure in this section is first written as data (the shape type, the list of items, the note holding its limit) and then drawn by the theme's diagram engine; the contract for that data is the _TEMPLATE.php file in this same plugin. What that order buys is exactly what a board gives a video team: the decisions are made on text, where changing them costs seconds, and because the figure is never drawn by hand, that one piece of data comes out right in four languages and in both reading directions. What it cost is equally clear: we only have shapes that have been coded, and any idea outside that list does not get drawn.

Real follow-up questions

How do I board if I cannot draw?

With rectangles and stick figures. A rectangle is the frame, a circle is a head, an arrow shows the direction of movement, and that is enough. If you are still uncomfortable, photograph instead of drawing: with a phone, at the location or anything resembling it, and put those photos beside the notes.

How many frames does a board for a thirty-second video have?

As many as there are shots, not a fixed number. The practical rule: one frame per shot, plus one extra for every shot where something enters or leaves the frame. If your frame count passes your second count, you are probably drawing the movement frame by frame, and that is no longer a board.

What is the difference between a storyboard and an animatic?

An animatic is the board's frames laid on a timeline with real durations and a temporary soundtrack. It shows the one thing a board cannot: pacing. If the video is expensive or its timing is tight, a simple animatic is worth the time it takes, because it reveals which shot looked fine on paper and drags in real time.

Keep learning