The Publishers Surviving the AI Traffic Collapse Are The Ones With Vertical Data Nobody Else Has

AI platforms like ChatGPT, Gemini, and Claude are answering the questions that used to send users to publisher websites. The result is a measurable, accelerating drop in search-driven traffic, which is showing consistent double-digit declines depending on the category. For publishers, that means fewer page views and less ad revenue. For the ad tech companies sitting between publishers and advertisers, it means fewer requests to broker. And for advertisers, it means less scale. The entire supply chain is feeling pressure from the same inflection point.

Helping publishers navigate the changing landscape is Alex Distadio, Vice President of Global Publisher Development at Doceree. Doceree is the world's first AI-powered operating system for healthcare marketing, unifying the prescription journey from physician awareness to fill on a single platform. Distadio's career spans nearly two decades in digital advertising, with senior roles at Taboola, Oracle, and Microsoft across North America, Latin America, and Europe. He sees the same revenue-model fight from 2008 playing out again, only faster.

"If a publisher has materially less traffic, it directly impacts revenue potential, and the ad tech plugged into that publisher has fewer requests to show advertisers. The whole market is being impacted." From his view, the publishers and advertisers who operate as content engines rather than single-channel destinations are best positioned to adapt.

We've seen this movie before

Distadio draws a direct line to the newspaper-to-internet transition he witnessed early in his career. In the late aughts, publishers migrated from print to digital and immediately faced the same tension: open everything up for maximum traffic and ad revenue, or gate content behind logins and subscriptions for fewer page views but deeper audience data and subscriber revenue. The industry debated it for years, and many publishers still haven't resolved it. "I see exactly the same thing today. To protect their websites against crawlers and AI robots, publishers are discussing whether to close the content. It means fewer page views, but more loyalty. And if you have a login or a paywall, you know the user's name, their location, and what they want. You're building first-party data, and first-party data is fundamental for targeting."

The parallel is instructive because as Distadio recalls, the publishers who survived the last transition were the ones who started defining themselves by their function instead of their format. "I had a manager 17 years ago who explained to me that newspapers, magazines, and portals were struggling because they didn't see themselves as a factory of content. The way they distribute it doesn't matter," he explains, noting the parallels in today's environment. "If you produce articles, you can monetize through apps, websites, social media, AI. You need to discover new revenue streams based on where user behavior is going."

The revenue streams publishers need now

Distadio sees three distinct paths publishers must build simultaneously to survive the traffic collapse.

The first is on-site LLM integration. Instead of losing users to ChatGPT, publishers can embed AI-powered chat directly into their own sites, allowing visitors to ask questions and receive answers drawn from the publisher's own content archive. The user stays on-site longer and engages more deeply, and the publisher retains the opportunity to serve contextual advertising against those interactions. "If you have AI integrated in your website, the user doesn't need to leave to go to ChatGPT. Depending on the question they ask, you can show related ads that make sense for that specific query."

The second stream is content marketplaces. If AI platforms are going to use publisher content to generate answers, publishers are increasingly pushing for compensation models. Distadio describes a marketplace model where publishers make their content available in a licensed environment and AI platforms pay for access rather than crawling it for free. "There is a movement in the industry, both in the US and Europe, to coordinate between AI platforms and publishers. The intention is to figure out ways to compensate publishers. If platforms rely on that content, sustainable compensation models will need to emerge," he shares.

The third revenue path is first-party data built through gated access. Whether through subscriptions, paywalls, or simple logins, publishers who know their audience can target more precisely and command higher CPMs. In healthcare, for example, Distadio points to NPI-level data, the unique ten-digit identifier assigned to every physician, as the kind of deep audience intelligence that makes a publisher's inventory irreplaceable. "Every doctor has an NPI number, and that number tells you their specialty. With that data, we can reach the right audience at the right time. We don't rely on cookies."

Vertical data is the moat

The ad tech companies most exposed to the traffic collapse are the horizontal middlemen, platforms that aggregate undifferentiated inventory across the open web and broker it to advertisers without proprietary audience data. When traffic drops 20%, their entire value proposition shrinks by the same margin. By contrast, the companies that are insulated are the ones with deep vertical data that advertisers can't get anywhere else. Distadio's own platform operates exclusively in healthcare, connecting pharmaceutical advertisers with verified physicians across a network of medical publishers. That specificity makes the inventory defensible even as broader open-web traffic contracts. "We have a very strong first-party data set because we're a company with digital DNA for healthcare," he says. "The founder was a doctor who built a digital company. We know the needs of doctors and healthcare providers. That explains the company's success, because we don't depend on cookies. We can target precisely."

For ad tech sales organizations in any vertical, the lesson is that generalist scale is becoming a liability. The defensible position is knowing your audience better than anyone else and building data assets that can't be replicated by a crawling algorithm.

Follow the user, not the channel

Distadio closes with a story that doubles as a GTM philosophy. At a conference in Florida, his manager asked whether he was heading to the pool after the sessions ended. He recognizes that since his clients were at the pool, that's where he needed to be. "That's exactly what publishers and advertisers need to do. There's a new behavior. The way you use the internet is different from the way I do. You need to go where your audience is, not where your channel used to be." The publishers and ad tech companies that survive this transition recognize that content is the product, distribution is variable, and the only durable advantage is knowing your audience well enough to follow them wherever they lead.