The Two-Track Web Is Here: What Publishers Need to Do Now

Key Points

• The Economist is building parallel content structures: rich human-facing pages and stripped-back, Q&A-formatted versions optimized for AI agents.
• Agent optimization is becoming a baseline requirement, not a differentiator. Publishers who skip it risk technical invisibility as AI reshapes discovery.
• Discoverability through agents doesn't replace the trust that drives subscriptions and premium ad revenue. It's a separate problem.
• Publishers need to decide which content earns agent-readable treatment and how much they can expose without eroding their core value proposition.
• The traffic you still control today needs to work harder than ever, which means your monetization setup has to be sharp.

According to Digiday's reporting on The Economist's AI strategy, the publication is actively building two versions of its web presence. One is optimized for human readers. The other is structured for AI agents: clean text, Q&A formats, no carousels, no feature art. Josh Muncke, VP of Generative AI at The Economist Group, calls it preparing for "a world with two versions of the web."

That framing is useful. It gives publishers a mental model for a problem that's been creeping up for months.

What The Economist Is Actually Doing

The Economist's current experiments are deliberately scoped. The team is restructuring content that already sits outside the paywall: marketing copy, B2B sales material, and pages designed for discoverability. The logic behind the move: a growing share of B2B buyers now start their research in ChatGPT, Gemini, or Claude. If your sales and marketing pages don't surface cleanly in those answers, you're not in the conversation.

The approach means building two versions of the same pitch. Humans get comparison-heavy, visually rich pages. Agents get stripped-back Q&A structures. Muncke describes the current work as "first, tentative experiments," with internal sandboxes used to test accuracy, tone, and performance before anything goes wider.

Critically, the editorial content behind the paywall is a separate, harder question. The Economist hasn't solved it. They're thinking carefully about it, which is the correct posture.

See It In Action:

• Lifestyle, Health & Travel Ad Revenue Resource Center: How publishers in traffic-sensitive verticals are protecting and growing revenue as discovery shifts
• News Publishers Ad Revenue Resource Center: Real approaches news publishers are using to maintain ad revenue as AI intermediation grows
• Entertainment Website Ad Revenue Guide: Monetization strategies for entertainment publishers facing changing traffic and engagement patterns

Why This Matters for Ad-Supported Publishers

The Economist is a subscription publisher. Their primary monetization risk is subscription erosion, not CPM dilution. Ad-supported publishers face a different version of the same problem.

If AI agents become the primary discovery layer, and they increasingly are, the traffic that reaches your site has already been filtered and summarized. Users arrive with less context to explore. Session depth shrinks. Pageviews per session drop. Every one of those metrics hits your ad revenue directly.

Alessandro De Zanche, founder of media consultancy ADZ Strategies, put it cleanly in the Digiday piece: "Agent optimization is a defensive baseline. Every quality publisher will build some version: the alternative is technical invisibility as search rebuilds around agents." He's right. But he also flagged the harder problem: "Agents drive discovery, not the trust and engagement subscriptions and premium advertising depend on."

That gap is where publishers need to think carefully. Agent-optimized content gets you found. It doesn't get you paid.

Essential Background Reading:

• AI and Publishers Resource Center: Everything publishers need to understand about how AI is reshaping content discovery, traffic, and monetization
• AI Crawler Resource Center for Publishers: Practical guidance on understanding and managing AI crawler exposure across your content
• Publisher Ad Revenue Maturity Model: A framework for assessing where your monetization setup stands and what to prioritize next

The Two Decisions Publishers Need to Make

This isn't a single decision. It's two separate ones, and conflating them causes problems.

The first decision is about discoverability: whether to structure content for AI agents at all, and which content earns that treatment. For most ad-supported publishers, the answer to the first part is yes. Marketing pages, evergreen content, and high-intent topic pages are obvious candidates. Gating all of it from agents is a visibility gamble that mostly hurts you.

The second decision is about protection: which content to block from AI crawlers, and how aggressively. This is where your editorial content, your differentiated reporting, and your unique data live. Giving it away to training pipelines without compensation is a different problem from optimizing for agent discoverability. Publishers are increasingly treating these as separate tracks, because they are.

Publishers working through this can use our AI Crawler Protection Grader to assess their current exposure and our AI crawler resource center for more detailed guidance on both tracks.

Related Content:

• Generative AI and Publishers: How generative AI is changing the content and advertising landscape for digital publishers
• Publisher Ad Tech Stack: What a modern, well-optimized ad tech stack looks like and how it supports revenue in a shifting traffic environment
• Yield Experiment Playbook: A structured approach to testing and improving yield when every session has to work harder
• News Publisher Guide: Monetization strategies tailored to news publishers navigating traffic volatility and platform dependence

What Publishers Should Do Now

The practical steps here aren't complicated, but they require deliberate prioritization.

• Agent-optimize your marketing and evergreen content: structure it as Q&A, use clear headers, strip unnecessary visual complexity. This is table stakes for discoverability.
• Audit your crawlable surface: know exactly which content is accessible to AI agents and which isn't. Most publishers don't have a clear picture of this right now.
• Separate the discoverability decision from the protection decision: one is about marketing and top-of-funnel discovery, the other is about protecting editorial value. Treat them differently.
• Maximize revenue from the traffic you still control: if agent intermediation is shrinking your top-of-funnel, your on-site monetization has to compensate. Yield optimization, session-level RPS, and ad layout quality all matter more in a compressed-traffic environment.

The Economist's Muncke was clear that nobody wants to read an AI-written Economist. The same logic applies in reverse: nobody is building an AI agent that recommends a publisher whose content can't be found or understood by the agent layer.

Next Steps:

• Publisher Ad Revenue Maturity Model Assessment: Take the assessment to identify gaps in your monetization setup before compressed traffic exposes them
• Publisher Earnings Index: Benchmark your revenue performance against publishers operating in the same environment
• Publisher Ad Revenue Maturity Model Guide: Step-by-step guidance for advancing your monetization operations as traffic patterns shift

The Monetization Layer That Can't Be Outsourced to Agents

Here's the part most of the industry conversation skips. Agent optimization is a distribution problem. Monetization is an execution problem. Getting discovered is different from converting that discovery into revenue.

Publishers who arrive at this moment with sloppy ad setups, poor viewability, or underoptimized yield configurations are going to feel the squeeze faster than publishers who've tightened their operations. Compressed traffic means every session has to work harder.

We work with publishers across gaming, entertainment, education, sports, and news on exactly this. Our RAMP platform handles yield optimization, demand access, and real-time auction mechanics that make each session count. When traffic gets more competitive for attention, the monetization infrastructure underneath it matters more, not less.

The two-track web is arriving whether publishers are ready or not. The Economist is getting ready. The question is whether you are too.