Which of the following is considered a best practice when creating effective image assets?

Analysis of Correct Answer(s)

• Always upload four or more relevant and unique images at the campaign or ad group level. This is considered a best practice because modern advertising platforms, especially those leveraging AI technology for performance optimization, thrive on a diversity of assets. Providing multiple (e.g., four or more) unique and relevant images allows the AI to:
  • Test various combinations: The system can experiment with different images alongside various headlines and descriptions.
  • Learn what resonates: AI identifies which images perform best with specific audience segments, across different placements, and in varying contexts.
  • Optimize dynamically: It can automatically serve the highest-performing image or combination, leading to improved ad effectiveness and better return on ad spend.
  • Maximize reach: A variety of images ensures the ad can adapt to different ad formats and placements, increasing serving opportunities.

Analysis of Incorrect Options

• Keep the most important content in the center 50% of the image asset. While good design practice often suggests a safe zone for critical elements to prevent cropping, a rigid "center 50%" rule is not the primary best practice for creating effective image assets for AI-driven systems. AI often resizes and crops dynamically, so a well-composed image that works across different aspect ratios is generally preferred. More importantly, this advice doesn't address the need for asset volume and variety that AI systems require for optimization.

• Only provide square images to maximize how often assets can serve. This is incorrect. Limiting assets to only square images significantly restricts serving opportunities because many ad placements (e.g., horizontal banners, vertical stories/reels) require or perform better with different aspect ratios. To maximize how often assets can serve, it's best to provide images in multiple aspect ratios (e.g., landscape, square, vertical) to allow the AI to adapt the creative to the most suitable format for each placement and user experience.

• Choose images with a white background as they tend to have stronger performance. This is a generalization that lacks universal validity. While white backgrounds can sometimes make a product stand out, there is no consistent evidence that they always lead to stronger performance across all industries, products, and target audiences. Performance is highly context-dependent. Lifestyle images, contextual images, or images with varied backgrounds can often perform just as well, or even better, depending on the campaign goals. Relying on AI's testing capabilities with a variety of image styles (not just white backgrounds) is a more effective strategy for discovering optimal performance.