Skip to content
Go back

Audience 3.0: The Great Divergence

7 min read

For two years, I’ve been talking about Audience 3.0 in every room I enter. Conference keynotes. Board rooms. Webinars with hundreds of digital publishers. Small off-the-record sessions with media executives trying to understand why their products feel increasingly out of sync with reality.

Our audiences have moved on. Our organizations have not.

The pattern is devastatingly simple: It’s as if our audiences have passed through a portal to an entirely different reality and now we’re broadcasting to empty space.

Our audiences have moved on. Our organizations have not.

This post is my attempt to write down what I mean by Audience 3.0 and why it matters so much for anyone who builds products, content or companies in media.

The Three Generations

Media audiences haven’t evolved into different species—they’re the same people who have chosen to live in fundamentally different realities. Three distinct behavioral modes now coexist, and most of us move between them throughout our day:

Audience 1.0: The Broadcast Generation These are spectators in a scheduled world. They consume what we decide to show them, when we decide to show it. One newspaper for the whole city. One evening newscast for everyone. The same magazine cover at every newsstand. We controlled distribution, they controlled the remote.

Audience 2.0: The Digital Transition The logged-in users of the early internet age. They click through websites and apps, follow topics and outlets, expect on-demand access. We built paywalls, recommendation engines, and newsletter strategies for them. They gave us email addresses and credit cards in exchange for some modest control over their experience.

Audience 3.0: Creator-Led, Feed-First, LLM-Fluent Here’s where the physics of media consumption fundamentally changed. This audience:

They are creator-led, feed-first, and LLM-fluent. And increasingly, they are everyone—regardless of age.

Recognizing Audience 3.0 in the Wild

Watch your own children. Watch yourself, if you’re honest. The behavioral patterns are unmistakable:

They don’t “visit” news sites—news finds them in feeds or they summon it through prompts. They don’t distinguish between consuming journalism versus entertainment versus education—it all flows through the same interfaces. A two-minute vertical video explains Israeli politics. A creator they trust breaks down inflation. ChatGPT summarizes the election results.

The traditional boundaries we built our entire industry around—sections, beats, formats, schedules—are meaningless abstractions to them.

When TikTok knows you’re interested in climate science, it doesn’t just show you climate content. It shows you climate content from creators you’ll trust, at the complexity level you can handle, in the format you prefer, at the moment you’re most receptive. Then it watches your response and adjusts.

That’s not personalization. That’s metamorphosis.

The Industry’s Defensive Crouch

In too many conference rooms, I hear the same exhausted refrains: This audience is shallow. They have goldfish attention spans. They lack critical thinking. TikTok is digital junk food. Real readers still come to homepages.

This is comfortable self-delusion.

They haven’t abandoned depth. They’ve abandoned our containers.

The Audience 3.0 cohort isn't shallow—they're navigating information at unprecedented velocity. They'll watch a three-hour video essay about Roman history from a YouTuber, then context-switch to a 30-second breaking news clip, then dive into a Reddit thread for nuanced discussion. They fact-check through parallel sources, not single authorities. They can smell institutional BS from a thousand yards away.

They haven’t abandoned depth. They’ve abandoned our containers.

The Structural Impossibility

Here’s what makes this genuinely existential: Most media organizations are architecturally incapable of serving Audience 3.0.

Our CMSes publish articles. Our workflows assume every story has one canonical version. Our metrics measure pageviews and time on site—artifacts from when we controlled distribution. Our org charts divide digital from print from broadcast, as if those distinctions still map to audience reality.

We’ve spent the last decade making our Audience 2.0 products incrementally better, while our actual humans migrated to Audience 3.0 platforms. We optimized the wrong machine.

I know because I helped build the operating systems that run this industry. As CTO of Arc XP at The Washington Post—and previously in product and technology leadership for AP’s B2B business—I’ve overseen the SaaS platforms that power newsrooms, broadcast operations, and digital publishing for hundreds of the world’s leading media brands. Arc XP and AP’s platforms aren’t just internal tools; they’re the software infrastructure that CNN, BBC, Reuters, and countless others rely on to create and distribute their content.

Those systems made existing workflows more efficient. They helped newsrooms publish faster, coordinate better, and operate more smoothly. But they were built to optimize the media machine as it existed, not to question whether that machine still made sense.

What Audience 3.0 Actually Wants

Strip away the platform mechanics and technological complexity. This audience is asking for something surprisingly simple: They want stories that understand them as individuals, delivered in formats native to their actual lives.

They want:

They don’t hate journalism. They hate how we package it.

The Uncomfortable Implications

If we accept that Audience 3.0 represents the future—and demographics guarantee it does—then several uncomfortable truths follow:

They don’t hate journalism. They hate how we package it.

  1. Our core products are misaligned. Articles, sections, and homepages are artifacts from a broadcast world.

  2. Our brands mean less than our creators. People follow Taylor Lorenz, not The Washington Post. Casey Newton, not The Verge.

  3. Distribution is now outside our control. We can either figure out how to thrive in feeds and chats, or we can become irrelevant.

  4. One-size-fits-all content is dead. The same story needs to exist in radically different forms for different contexts.

  5. LLMs will intermediate our relationships. Our content will increasingly be consumed through AI interfaces we don’t control.

These aren’t trends to resist. They’re realities to design for.

The Path Forward

I’m not suggesting we chase every platform or abandon journalistic principles. I’m suggesting we recognize that our audiences have already moved and build products that meet them where they live.

This means:

Some organizations get this. The New York Times’ games and cooking apps acknowledge that not everything needs to be an article.

These aren’t trends to resist. They’re realities to design for.

The Information's Discord channels recognize that community matters as much as content. Individual journalists building massive followings on Substack and YouTube understand that trust is personal, not institutional.

But these are exceptions. The industry’s mainstream remains structured for a world that no longer exists.

Why This Matters Now

Every day we delay adapting to Audience 3.0 reality, we cede more ground to creators, platforms, and AI systems that have no such hesitation. Our audiences aren’t waiting for us to figure this out. They’ve already passed through the portal.

The question isn’t whether to serve Audience 3.0. It’s whether we’ll have any audience left by the time we try.

The question isn’t whether to serve Audience 3.0. It’s whether we’ll have any audience left by the time we try.

I believe in journalism’s necessity. I chose this industry because I believe stories shape the world. But if those stories can’t reach people in the reality where they actually live, we’re just talking to ourselves in empty space.

The divergence is complete. The portal has closed behind them.

The only question left is whether we’ll build our own way through.


Share this post on:

Previous Post
The Portal Problem