



Creativity has moved back to center stage. In an ecosystem where audiences are fragmenting, costs are rising, and targeting is being reshaped, understanding what makes an ad perform is no longer a “nice to have.” It’s no coincidence that multiple sources agree a significant share of incremental impact can be explained by the creative itself.
The challenge is that, in practice, most marketing teams operate with a creative black box. They may have thousands of active or historical assets, yet very limited systematic visibility into which visual and audio elements are repeated, which ones are missing, and which ones are associated with stronger performance.
This is where Creative Intelligence comes in: an approach that combines automation, AI, and Marketing Science methodologies to transform every creative asset into a set of measurable, comparable, and optimizable attributes.
IN THIS ARTICLE
Creative Intelligence is the technology that enables teams to automatically download, organize, and analyze all creative assets (images and videos) running with investment on Meta. These assets are processed using AI to perform qualitative content analysis at scale, allowing teams to understand their components and connect them with media investment and performance metrics.
In other words, it transforms creative assets into structured data.
The first leap is operational: automating the collection of creative assets and eliminating manual work (folders, exports, hand labeling, small samples).
Then comes the analytical leap: using AI, each creative asset is processed to identify components that are typically not available as metadata within advertising platforms:
For video, the analysis can also be performed frame by frame to detect critical signals such as overall logo presence and logo visibility in the first seconds, music or voice-over, product focus, and the dominant emotion.
This enables something key: an individual diagnostic view of each asset, as well as a macro-level perspective across campaigns, time periods, funnel stages, formats, and more.
Having attributes is the starting point. The real difference lies in the interpretation framework.
Creative Intelligence relies on a structured framework to analyze attention drivers and communication dimensions that are relevant for a brand, such as:
Instead of evaluating creatives “by eye” or based on subjective preference, the focus shifts to the ability to capture attention, communicate quickly, and generate real impact.
One of the biggest challenges in creative analysis is avoiding the pendulum between “creative opinionism” (purely subjective feedback) and overly atomized analysis (focusing on a single metric or detail while losing sight of the whole).
To address this, Creative Intelligence builds a Brilliant Basics Scoring for every asset: a set of 10 proprietary indices based on empirical evidence and platform learnings.
The logic is simple: standardize the evaluation of key components to answer questions such as:

A good practice for leveraging Creative Intelligence is to work across three layers:
Start by understanding the overall state of the creative ecosystem:
This makes it possible to identify gaps and opportunities that trigger more specific deep dives.
Using filters (for example, campaign_id), it is possible to audit:
Once you have structured attributes, you can run correlation analyses to build a hypothesis backlog, for example:
From there, these insights can be taken into experimentation (A/B tests, variant rotation, and increased creative diversity).
In marketing, the inevitable question is: does this actually cause the outcome, or is it simply associated with it?
Beyond descriptive analysis and visualization, Creative Intelligence can be complemented with dedicated Data Science projects to model the causal relationship between creative attributes (objects, colors, logo presence, emotions, etc.) and a target metric (CTR, CVR, CPL), while controlling for factors such as bidding strategy, budget, audience, and placement.
When Creative Intelligence is properly implemented, it changes the workflow:
In the age of automation, creativity doesn’t lose value, it gains responsibility. The brands that lead in creativity and innovation are those that manage to systematize creative learning and turn it into a competitive advantage.
At Bunker, we believe brands must preserve control over their strategic levers, avoiding the full delegation of these decisions to AI systems designed to optimize media or generate content.
Creative Intelligence proposes exactly that: turning creativity into data, data into insights, and insights into a roadmap for optimization and experimentation.
If you want creativity to stop being a black box, the path isn’t to look at more media dashboards, it’s to better understand what each asset is communicating and how that translates into attention and performance.

Lucas Suarez
Marketing Analyst
1/9