.svg)
TL;DR: Media consumption has moved on, but data practices are still stuck in the web era. Emerging channels like podcasting, connected TV and gaming show how identity-based targeting breaks down. With AI and on-device intelligence, we can finally move from identity to understanding, building sustainable, ID-less ways to connect with audiences across both ad-supported and subscription experiences.
The internet’s data problem
Data has always powered the internet and the advertising that funds much of it. It personalizes content, measures engagement, and drives better experiences. But while media consumption has evolved across mobile, connected TV, smart speakers, and gaming platforms, the industry’s data practices are still stuck in the web era.
Most systems still rely on identity, using cookies, device IDs, hashed emails, and clean rooms, assuming it’s the only way to deliver relevance. That worked when the web revolved around logins and browsers, but it doesn’t fit a world that’s app-based, cross-device, and privacy-first.
To see how we got here, it’s worth looking back at how data solutions in advertising have evolved and why each generation fixed one problem but created another.
- Data Management Platforms (DMPs): 2008 – 2017
- Why it emerged: the big data era, where the idea was to simply collect everything for 1-1 marketing.
- Core weaknesses: cookie-dependent, web-only, focused on collection, not insight
- Customer Data Platforms (CDPs): 2014 – present
- Why it emerged: increasing mobile usage led to data fragmentation, driving the need to unify 1st-party data across apps and websites.
- Core weaknesses: great for retention, but not for cross-platform activation or audience intelligence
- Identity Graphs / Alternative IDs: 2018 – present
- Why it emerged: attempts to rebuild identity across channels using hashed emails and logins.
- Core weaknesses: still identity-based; fails in anonymous or shared environments
- Clean Rooms: 2020 – present
- Why it emerged: privacy regulation required controlled, permissioned data sharing.
- Core weaknesses: scales poorly, still relies on matched IDs
Each of these approaches made sense for its time, but none anticipated how much the internet itself would evolve. Today, consumption happens across devices and environments where identifiers either don’t exist or don’t mean much. Think gaming, podcasting, CTV, shared devices—trying to reach individuals on these channels through hashed emails or shared logins has become closer to guesswork than precision, limiting both accuracy and scale.
The issue isn’t just privacy regulation or signal loss. It’s structural. We need a foundation that understands what people care about, not who they are.
From identity to intelligence
Three things have converged to make this shift inevitable.
First, privacy laws like GDPR (General Data Protection Regulation), CCPA (California Consumer Privacy Act) and the DMA (Digital Markets Act) have redefined what can be collected and shared.
Second, consumers are more aware of how their data is used and actively choose products that don’t track them.
And third, technology has caught up. AI and on-device computing now allow us to interpret behavior and intent without centralizing data and storing identifiers.
Instead of relying on identifiers to tell us who someone is, we can now understand what they’re doing and why. This marks a fundamental change, from identity to intent, from collection to understanding
The rise of ID-less, AI-powered intelligence
For years, the industry’s response to data loss has been to patch identity: new IDs, hashed logins, cleaner graphs. But what if the answer isn’t rebuilding identity at all?
ID-less data takes a different path. Instead of trying to recognize users, it uses machine learning to interpret intent, and behavior. Edge AI processes the data locally on devices, ensuring privacy by design. Large language models (LLMs) and natural language processing (NLP) systems can interpret unstructured signals such as text, audio, or app metadata and give them meaning.
These models easily scale across markets without heavy engineering or retraining, making them efficient and compliant. This isn’t another targeting trick. It’s a shift away from the mass-market obsession with collecting personal data toward a more sustainable, privacy-conscious approach to make sense of digital behavior.
You can already see this shift play out in podcasting, gaming and CTV, three of the fastest-growing channels in digital media. Each attracts massive audiences but faces the same structural problem: monetization built on identity doesn’t work when identity isn’t reliable. Podcast platforms hold fragmented listener data that’s hard to unify. Gaming delivers incredible engagement but offers few consistent signals to connect advertisers and players. CTV reaches millions of households, yet identity resolution remains messy and inconsistent. Traditional data systems weren’t designed for a world where consumption is fluid and privacy is the norm.
The only scalable way forward is to understand behavior and intent in privacy-safe ways, using intelligence instead of identity.
Rebuilding the internet’s data foundation
At NumberEight, we’re building toward that foundation: privacy by design, technology that adapts to how people actually consume content, and data that explains moments, not individuals.
Just a few years ago, “ID-less data” wasn’t even part of the industry conversation. Today, it’s being recognized as a practical path forward, reflected in new standards (e.g., ID-less Solutions Guidance, ID-less Solutions Landscape), agency strategies, and growing adoption across emerging channels.
This isn’t about replacing one identifier with another. It’s about moving away from the endless chase for personally identifiable information (PII), a losing battle, toward a holistic model that reflects how the internet truly works. It’s about moving from a data collection mindset to using intelligence to drive real relevance.
.webp)
.webp)
