Taxonomies Alone Won't Save Digital Advertising

"The digital advertising industry has always been eager to create standards that simplify complexity. Taxonomies-structured systems for labeling content and products-are one such attempt. And while the IAB Tech Lab's new guidance to connect Content Taxonomy 2.1 with Ad Product Taxonomy 2.0 represents progress, it also raises a fundamental question: Is this really the evolution we need? Or is it just a neater version of a system that no longer fits the reality of how people engage with content?"

"To be fair, taxonomies do serve a purpose. They bring consistency to a fragmented ecosystem. But as we stretch these frameworks across emerging formats like CTV, audio and short-form video, their limitations become harder to ignore. Static labels, however meticulously mapped, will always struggle to reflect the dynamic nature of human culture. And advertising, after all, is a business rooted in cultural relevance."

"Take "news," for example, as a category. If you label an ad-safe article about ICE enforcement on NPR and a sensationalized segment on the same topic from a cable opinion show as equivalent, you lose not just nuance but meaning. Context is never just about topic; it's about tone, audience and emotional resonance. And if a taxonomy can't differentiate between outrage and empathy, what are we really optimizing for?"

"AI Exacerbates The Issue This problem becomes even more urgent as we move toward environments that lean harder on AI-driven solutions. The irony is that while AI is supposedly more advanced than simple labels, most current implementations are just taxonomies in disguise. They repackage the same legacy structures with marginal improvements and rely on black-box modeling to suggest intelligence. But how many advertisers can actually validate what audiences they're reaching? How many understand how their "custom" segments were generated?"

Digital advertising has long used taxonomies to simplify and standardize content and product labeling. Efforts to align Content Taxonomy 2.1 with Ad Product Taxonomy 2.0 aim to improve interoperability but may not match modern content dynamics. Taxonomies provide consistency across fragmented ecosystems but struggle with emerging formats like CTV, audio, and short-form video. Static labels cannot reliably capture tone, audience, or emotional resonance, reducing contextual nuance. AI-driven solutions risk compounding these shortcomings by repackaging legacy taxonomy logic into opaque models. Many advertisers lack the ability to validate audience composition or understand how custom segments are constructed.

Read at Forbes