Variable Video Composer for Personalized Content Assembly
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Marketing companies face inefficiencies in producing personalized videos for targeted marketing, requiring massive manual replication and error-prone personalization to achieve variability across multiple videos.
Innovation Solution
A system comprising a composer tool for building and a player tool for dynamically assembling variable videos in real-time from component media elements, using pre-buffering and a cloud-based database to deliver personalized content based on viewer metadata and responses, allowing for scalable, automated, and cost-effective production and delivery of personalized videos.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If discrete sets of versioned videos are produced manually, then personalized messaging can be achieved, but production complexity and manual labor increase significantly
Solution Approach 1:
The video is divided into modular components including variable acts, scenes, and media elements that can be independently selected and assembled. This segmentation allows the system to generate personalized videos by combining different modules rather than manually producing each variant, thereby reducing production complexity while maintaining personalization capability.
Solution Approach 2:
The system transitions from static pre-produced video variants to dynamic real-time assembly of videos based on viewer metadata and responses. The player tool dynamically selects and assembles variable video components during playback, enabling personalization without requiring manual production of each customized version.
2Adaptability or versatility
If manual personalization is performed for each video variant, then relevant variability can be achieved, but time consumption and error rates increase
Solution Approach 1:
Video components such as acts, scenes, and media elements are pre-produced and stored in a library before actual personalization is needed. The system uses these pre-prepared components to generate personalized videos on demand, eliminating the need for time-consuming manual personalization of each variant while maintaining video variability.
Solution Approach 2:
Instead of manually creating each personalized video from scratch, the system uses templates and component libraries to generate video variants by copying and recombining existing media elements. This approach maintains video variability while significantly reducing production time compared to manual personalization of each variant.
3Adaptability or versatility
If multiple discrete video versions are produced, then targeted marketing can be achieved, but production costs increase
Solution Approach 1:
A single variable video file contains multiple versions and variants that can be served to different audiences based on their metadata. This universal approach allows one production to serve multiple marketing targets simultaneously, reducing production costs compared to creating separate discrete video versions for each target segment.
Solution Approach 2:
The system changes parameters such as media selection, scene sequencing, and content rendering based on viewer metadata rather than producing entirely different video files. This parameter-based approach maintains targeted marketing capability while reducing production costs by reusing the same base video structure and media libraries across different versions.
Data Source
AI summary
An example composer to author a variable video includes a builder to create scenes, a content retriever to assign content to the scenes, a library interface to point to the content for the scenes, a labeler to tag the scenes with at least one vector based on the content, a receiver to obtain first relevance data from first intended viewer of a first version of the variable video and to obtain second relevance data from a second intended viewer of a second version of the variable video, a mapper to chart a first sequence of two or more scenes based on a vector and the first relevance data and to chart a second sequence of two or more scenes based on a vector and the second relevance data, and a publisher to publish the variable video as a single file based on the first sequence and the second sequence.