MakeUGC is an AI platform where brands create video ads with virtual actors. The product worked, but it was fragmented into too many separate actions. The redesign reorganized it around a single creation flow.
Context
MakeUGC generates UGC-style video ads: AI actors, voices, scripts, b-roll, and product staging. Features were multiplying fast, and each capability lived as an independent tool with its own entry point.
The core problem was the workflow. The user had to assemble the process in their head, and every new surface became another place to stop.
Prompt-led creation
The core decision was to make a single prompt bar the entry point of the product. The user describes the video, attaches product assets, and picks the generation mode directly in the bar: talking actors, b-roll, product-in-hand, or the video agent.
Script writing, voice, and text generation are one tap away instead of living on separate screens.
Workflow
For longer generations, the video agent drives the process as a conversation. It analyzes the input, proposes scenes, and asks for decisions along the way.
The agent asks questions with selectable answers, and a progress list makes the wait transparent instead of silent. Retention got its own flow: cancellation offers a pause and targeted deals before someone unsubscribes.
Website
The homepage needed to explain the offering quickly, without making the product feel like a stack of disconnected AI tools.
Pricing
Pricing translates abstract credits into concrete outcomes — generations and videos — so plans compare themselves.
The real shift was turning a set of separate generation features into a single flow the user could follow without having to think about the product's structure.