Motion graphics and animation

Script, storyboard, styleframe: why animation is decided before it is animated

Animation is expensive to change and cheap to plan, so almost all the important decisions happen before anyone opens the animation software. The sequence is script, then storyboard, then styleframes, then animatic, then animation.

The script comes first because animation length is set by the voiceover, not the other way around. Roughly 140 to 160 spoken words fills a minute at a comfortable pace. If a client wants a 60 second explainer, that is the budget, and cutting the script later means re-recording and re-timing everything downstream. We usually record a temporary voiceover as soon as the script is approved, so every later stage is timed against real speech.

The storyboard is the sequence of key moments as rough drawings. It settles what appears on screen and in what order. Styleframes come next: two or three fully designed frames pulled from the storyboard, rendered at final quality. These are the approval point for colour, type, illustration style and level of detail. Signing off styleframes rather than a finished animation is what keeps the project affordable, because changing a still is minutes of work and changing a finished sequence is days.

• Script first, roughly 150 words per finished minute
• Scratch voiceover recorded early so timing is real, not guessed
• Storyboard settles what happens and in what order
• Styleframes settle how it looks, at full final quality
• Animatic times the storyboard to the voiceover before animation starts

What makes animation feel expensive: easing, timing and secondary motion

The difference between animation that looks professional and animation that looks like a slide transition comes down to how things start and stop moving. Nothing in the physical world begins at full speed and stops instantly, and motion that does reads as cheap even to viewers who could not say why.

Easing is the curve applied to a movement. Ease-out, where an object starts fast and settles slowly, suits most interface and text motion because it feels responsive. Ease-in-out suits camera moves and larger objects. Linear motion, with no easing at all, is the signature of unconsidered work and should be reserved for things that genuinely move mechanically, like a loading bar.

Timing is the other half. Most motion graphics elements should complete their move in roughly a quarter to half a second. Faster feels twitchy, slower feels sluggish. Secondary motion is what sells it: a slight overshoot and settle, a shadow that lags the object, elements that arrive staggered rather than all at once. Staggering a list of five items by a few frames each costs nothing and immediately looks considered.

Choosing between 2D motion, character animation and 3D

Three broad approaches cover almost every brief, and they differ enormously in cost and revision flexibility. Picking the wrong one is the most expensive mistake in an animation project.

2D motion graphics moves shapes, icons, type and illustration. It is the workhorse for explainers, product walkthroughs, social animation and data visuals. It is fast to revise, sits naturally alongside a brand identity, and scales well because assets are reusable. Character animation adds people or mascots that move and express. It is warmer and better for stories about human situations, but it is significantly more labour, because every character needs rigging and every movement needs performance rather than just position changes.

3D gives physical realism, real lighting and camera moves through space. It is the right answer for products with genuine physical form, mechanical explanations, or anything where the viewer needs to understand how a thing is built. It is the slowest to change, because a note like "make it a bit warmer" can mean re-rendering every shot. When a brief says 3D but the goal is clarity rather than realism, a well-designed 2D approach usually explains it better and for less.

• 2D motion: fastest, most flexible, best for explaining concepts and processes
• Character animation: warmer and more relatable, notably more production time
• 3D: for physical products and mechanical detail, slowest to revise
• Mixed approach: 3D product shots inside a 2D graphic system is often the practical middle

Building a motion system rather than a one-off video

Brands that use animation regularly should own a motion system, not a collection of unrelated videos. A motion system defines how your brand moves, in the same way a style guide defines how it looks.

It covers a small number of decisions applied consistently. How does type arrive on screen: fade, slide, or a wipe from a mask? What easing curve is the house default? How long does a standard transition take? Which two or three transition types are allowed, so every video does not invent its own? How does the logo animate, and is there a short and long version? What is the lower third layout and how does it enter and leave?

Documented once, these turn every future project into assembly rather than invention. They also mean different animators produce work that looks like it came from the same brand. We deliver the system as a written reference plus editable project templates, so simple assets like title cards, subtitle animations, social frames and end cards can be produced by your own team without commissioning anything.

• A house easing curve and standard transition duration
• A defined set of allowed transitions, deliberately small
• Logo animation in short and long versions
• Lower third, title card and end card templates
• Editable project files so your team can produce routine assets

Animated data visuals that inform rather than mislead

Animating a chart is easy. Animating it honestly takes more thought. Motion draws attention to change, which means a badly animated chart can imply a story the numbers do not support.

The rules are straightforward. Bar charts start from a true zero baseline, because an axis that begins partway up exaggerates differences and animation makes the exaggeration more convincing. Reveal one series at a time so the viewer can absorb each before the next arrives. Hold each state on screen long enough to be read, which is usually longer than feels necessary while editing. A number counting up should end on the real figure and stay there for at least a second.

Labelling matters more in motion than in print, because the viewer cannot pause to check. Every axis and unit should be on screen while the data is. Sources should appear too, and if a figure came from your own internal reporting it should say so. Animation that looks authoritative while hiding where the numbers came from is a credibility problem, not a design one.

Formats, transparency and where the animation will be used

Where an animation will run changes how it must be built, and the decision has to be made before animation starts rather than at export. A logo sting for a video edit, a looping web header and an animated ad unit have almost nothing in common technically.

Anything that needs to sit over other footage or over a web page needs a transparent background, which rules out standard MP4. Those deliver as a format with an alpha channel, commonly ProRes 4444 or a WebM file for browser use. Anything for the web needs to be small: a Lottie file, which is vector animation played by code, can be a fraction of the size of a video and stays sharp at any resolution, though it suits flat vector work rather than anything with complex effects. Animated GIF still exists but is heavy and colour-limited, and is rarely the right answer now.

Length limits also shape the build. Ad platforms cap durations and file sizes, and some placements strip audio entirely, which means the animation has to work silently. We confirm the destination list at brief stage and design once for the strictest constraint, rather than animating freely and then discovering the piece cannot be delivered where it was needed.

• Transparent overlays need an alpha format, not MP4
• Lottie for lightweight vector animation on websites and apps
• Design for silent playback wherever the placement mutes audio
• Confirm duration and file size caps before animating, not at export