Generative AI Creative Layer Selection for Personalized Ads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional creative generation processes are time-consuming and inefficient, often resulting in generic advertisements that do not effectively resonate with individual audience members, leading to lost revenues and suboptimal targeting.
Innovation Solution
A system utilizing generative AI models to create customized creatives by analyzing digital content, identifying available layers, and combining them based on audience preferences and criteria, enabling on-the-fly generation and personalization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional manual creative generation is used, then creative content can be produced, but the process is time-consuming and costly
Solution Approach 1:
The patent replaces manual mechanical creative generation processes with an automated AI-based system. The creative engine uses machine learning models to automatically generate, modify, and optimize creative content based on campaign objectives and audience data, eliminating the need for time-consuming manual iterations while maintaining creative quality.
Solution Approach 2:
The system performs preliminary actions by pre-processing and organizing creative assets, audiences, and campaign parameters before actual creative generation is needed. The creative engine maintains ready-to-use creative libraries and pre-segmented audience data, enabling rapid deployment of customized creatives without time-consuming preparation during execution.
2Productivity
If traditional manual creative generation is used, then creative content can be produced, but it is costly in terms of resources and time
Solution Approach 1:
The patent replaces resource-intensive manual creative production with an automated AI system that consumes computational resources instead of human labor. The creative engine uses machine learning models to efficiently generate multiple creative variations simultaneously, reducing overall resource consumption compared to manual processes while producing higher quality output.
Solution Approach 2:
The creative engine performs self-service by automatically optimizing its own creative generation based on feedback from performance data. The system continuously learns from campaign results and automatically adjusts its generation parameters, eliminating the need for external human intervention and reducing overall resource consumption through automated optimization.
3Adaptability or versatility
If a singular generic advertisement is displayed to the target audience, then the advertisement can be delivered, but it does not resonate with individual audience members
Solution Approach 1:
The patent segments the target audience into distinct groups based on demographics, interests, and behavior patterns. The creative engine then generates customized creative variations for each segment, allowing precise targeting of individual audience members with relevant content rather than using a generic approach that fails to resonate with specific individuals.
Solution Approach 2:
The system applies local quality by customizing creative attributes such as imagery, copy, and video content to match the specific characteristics of individual audience members or small segments. Each creative delivery is optimized for the local context of the recipient, ensuring high relevance and resonance rather than using a one-size-fits-all generic advertisement.
4Productivity
If traditional creative generation processes are used, then creative content can be produced, but iterations and vetting are required before finalization
Solution Approach 1:
The patent replaces complex manual vetting and iteration processes with automated AI-based evaluation systems. The creative engine automatically assesses generated creatives against campaign objectives, audience data, and performance predictions, eliminating the need for human review iterations while maintaining high creative quality and reducing overall process complexity through automation.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Systems and methods for custom creative generation are provided herein. An electronic request to generate a custom creative from a digital content is received. Available creative layers for use in the custom creative are identified based at least in part upon one or more data tags associated with the digital content. The one or more data tags indicate creative features of the digital content. Machine learning is used to identify a subset of the available creative layers based upon generation criteria associated with the electronic request. The custom creative is generated by combining the subset of available creative layers. An electronic response is generated and provided in response to the electronic request. The electronic response includes an indication of the custom creative.