Mechanism-of-Action and Product Motion, 2D and 3D

A mechanism-of-action animation takes five to seven weeks, and the first three of those contain no animation at all. They go on a scientific storyboard that medical affairs signs frame by frame. The expensive mistake in this work is never a bad render. It is a beautiful render of a mechanism the approved product documentation does not describe. How we build: 2D runs in After Effects. 3D runs in Cinema 4D with Redshift. Simulation goes to Houdini when a fluid, a particle field or a soft body has to behave rather than be posed. Product and component geometry arrives from your engineering team as STEP or IGES and gets rebuilt for render rather than decimated, because a chamfer that reads wrong on a medical device is a factual error, not a shading choice. Web delivery goes out as Lottie where the asset sits inside an interface, since a 40 KB JSON that stays sharp at any density beats a looping MP4 in a page already fighting for its LCP, and as ProRes or H.265 where it is a film. The judgement you are buying: Choosing what not to show. Every molecule has ten mechanisms and one of them changes what a prescriber or a parent does next. We will spend a session arguing you down from four mechanisms to one, and that argument is the part that makes the animation work. The same holds in automotive. A cutaway showing the whole drivetrain shows nothing. Show the one component the buyer does not believe in. Rules we hold to inside these categories: Scale and speed carry a stated relationship to reality or an on-screen note that they do not. Effect sizes are not animated larger than the evidence supports. Cells do not turn a triumphant colour when the mechanism engages, because colour reads as efficacy to a viewer and nobody signs off on that reading. Device animations state model and configuration. Anything a child will watch stays under the flash thresholds in WCAG 2.3.1 and the harmonised broadcast guidance, where three flashes in any second is a limit rather than a style note. What we refuse: No template explainers. No kinetic typography over a stock library. No whiteboard-hand animation. We refuse to animate a claim that is not traceable to a named document: the approved product characteristics, the clinical publication, the homologation figure, the test protocol. We ask for that document by name before storyboard. We do not do character animation either. If the idea needs a character with a performance, you want an animation studio, and we will say so in the first call. Who should not hire us: Anyone who wants a video explaining what the company does in general. That brief has no failure condition and therefore no way to be right. Anyone who needs the animation before the science is settled internally, because we will stop and wait for it on your clock, and neither of us enjoys that month. How it runs: Discovery with your medical, regulatory or engineering lead in the room rather than relayed through marketing. Scientific storyboard, reviewed and marked up. Style frames in two directions, one chosen. Animatic with scratch voice-over and locked timing, which is where changes are cheap and where we push hard for them. Then animation, look development, grade, sound design and mix. Reviews run in Frame.io with time-coded comments, so a regulatory reviewer marks a frame instead of writing a paragraph about the middle bit. Delivery includes the master in your finishing format, a textless version, layered source for your in-house team if you have one, and the still frames the animation was built around, which usually end up in the deck and the detail aid within a month.