After Effects is a 1993 application with thirty years of sediment on it. Every serious motion designer has felt the moment where the tool is the obstacle. So a generation of replacements arrived. Here is where each one genuinely wins, from someone who has shipped paid work in all of them.
Cavalry — the one that should have won
Cavalry is what After Effects would be if it were designed now. Procedural by default, a real node-ish data model, duplicators that are not a plugin, and an interface that does not fight you.
For anything generative — data-driven charts, hundreds of instances, systems rather than scenes — Cavalry does in an afternoon what After Effects does in a week of expression debt.
Why it has not finished the job: the plugin ecosystem, and the fact that no client's post house accepts a .cv file. Cavalry renders and hands off; it does not sit inside anybody's existing pipeline.
Rive — the one actually replacing After Effects in product work
This is the real story, and it is under-reported. If your animation ships inside an app or a website, After Effects was always the wrong tool. You animated in AE, exported via Bodymovin/Lottie, and then spent a day discovering which of your effects Lottie silently drops. (It drops most of them.)
Rive skips that. You author the interactive thing itself, with a state machine, and it runs at runtime at a fraction of the file size. Hover states, loading states, buttons that respond — this is Rive's territory and After Effects has effectively conceded it.
Why it has not finished the job: it is not a film tool. Nobody is grading a commercial in Rive.
Jitter — the one that took the easy 80%
Jitter is a browser-based motion tool aimed squarely at product designers who need a decent animated mockup by Thursday and do not want to learn a compositor.
It is genuinely good at that, and honestly it took a real slice of low-end motion work with it. That work was never enjoyable and never paid well.
Why it has not finished the job: ceiling. You reach it fast.
Theatre.js — the one for people who would rather write code
A proper animation tool for the web, with a real timeline UI, driving actual JavaScript objects. If your output is a Three.js scene or a WebGL hero, this beats exporting a video of a 3D scene and pretending.
Why it has not finished the job: it is for developers. That is not a criticism, it is a scope.
So what actually killed what
Nothing killed After Effects. What happened is more interesting: After Effects lost the categories it was never good at.
- Interactive UI animation → Rive
- Generative and data-driven → Cavalry
- Quick product mockups → Jitter
- Web-native 3D → Theatre.js / code
- Edit, grade, finish → Resolve
What is left is what After Effects was always genuinely best at: bespoke, art-directed, frame-by-frame motion graphics with a deep effects pipeline. That is a smaller category than it was in 2015. It is also the category that still pays the most.
The practical advice
Do not switch. Add.
Learn Rive if you work with product teams — it will get you hired, because almost no motion designer can hand a developer a working state machine.
Learn Cavalry if you do data or systems work.
Keep After Effects, because when a client says "can we just try it with the camera pushing in and a bit more glow on the third one", every one of these alternatives becomes the reason you missed the deadline.
I build motion for SaaS and product teams, including the Lottie-and-Rive end where it has to run in the actual product. Send me a brief and I will tell you honestly which of these your job wants.