Traffic is dying as a media metric. What comes next is more important

Day by day there’s more evidence that AI is eating up the media world. A recent report from Growtika, a self-described SEO and AI search agency, analyzed data from the search analytics platform Ahrefs to show that traffic to many tech media sites is way down over the past couple of years.

Hardest hit were Digital Trends (down 97%), ZDNet (down 90%), and The Verge (down 85%). Even the most seemingly resilient publications (Mashable was down only 30% and CNET 47%, both Ziff-Davis properties) took significant hits. Some of these reductions are no doubt exaggerated—Growtika compared each publication’s peak month with traffic in January 2026, which doesn’t account for seasonal reductions—but no one’s disputing the overall trend, or who’s to blame: AI.

At first glance, the numbers appear to reinforce the idea that the bottom is rapidly falling out from beneath the media industry. But that’s overly simplistic, and it fails to take into account that publishers have seen this trend for years, and many have been adapting around it. The Verge is actually a great example of this: Not only was it out early in questioning what AI does to content, but it also introduced a paywall in late 2024, part of a larger, four-point strategy. So traffic decline doesn’t necessarily mean business decline.

Authority still wins

The Verge is an example of something else that’s important in the AI era: having a strong brand with a loyal audience. The publication has been synonymous with tech news, commentary, and analysis since its debut in 2011. Many tech brands choose to break their news there. That credibility has influence in what appears in AI answers, which tend to favor journalistic content above other types—a point that a recent Gartner report on the communications industry hammered home.

So while it’s easy to see AI as a traffic destroyer, the flip side of that is it’s an audience qualifier—the people who find your brand through an AI summary are clearly deeply engaged. In other words, they’re probably the most willing to pay for your content, and The Verge has given them that opportunity.

But The Verge has had a strong brand for 15 years. The audience is clear and already engaged. For many publications, that might not be as true, and the Growtika numbers should be a wake-up call for them to better understand their brand, their audience, and even what business they’re in. With AI now fully in the picture as the ultimate information synthesizer, publishers have to understand what they’re layering on top of that information that only they—specifically, the humans that work for them—can provide.

Journalists tell stories because humans are storytellers. More importantly, they’re story listeners—audiences will seek out publications and writers with voices that resonate with them, AI or no AI. The content that doesn’t do that, by contrast, has less value and is easily substituted by AI. For most publishers, this means favoring analysis, opinion, and scoops. Aggregated news (i.e. stuff broken elsewhere, or broad announcements) is less valuable, though with a caveat: It’s difficult for a publication to appear authoritative if they don’t cover “big” news on their beat. But for that to be worthwhile, it requires that:

1. You have a very clear idea of what your beat is.
2. When you cover it, you seize the opportunity to show your unique value.

Small is the new big

The implication of that approach is that it favors smaller publications and independent voices, and it’s no surprise that those areas are flourishing in the age of AI. Substack reached 5 million paid subscriptions last year, and competitor Beehiiv—home to many journalists who used to work at places like The Verge and CNN—is set to double its revenue this year to $50 million.

Still, some of the numbers in the Growtika report suggest that larger organizations with strong internet domains still have resilience. Mashable and CNET are both Ziff publications, and having worked for the company twice, I can say with some authority that Ziff is very strong when it comes to SEO. Larger outlets also have resources to quickly spin up formats that audiences are gravitating towards, such as vertical videos.

But ultimately, those resources will dwindle if they haven’t taken a hard look at their audience, the value they’re bringing, and how they’re encouraging loyalty (or “conversion” if you’re on the business side). For many publications, especially those that thrived during the 2010s, that means narrowing focus. It also means adapting or even pivoting your business around a new model. There’s a reason media events are on the rise; research from Vendelux, an event intelligence company, reported that 23% of publishers received a “large or very large portion” of revenue from events in Q1 2025, up from 8% in Q1 2024.

The metric after metrics

So what the great traffic drop of 2026 tells us isn’t that AI is eating the media, but traffic is losing its relevance as an indicator of success. Brands that have built businesses around things that aren’t just “articles”—such as subscriptions, newsletters, podcasts, and events—can still succeed, albeit with a somewhat altered definition of success. In almost all cases, the audience is fewer in number, but more engaged in a way that’s measurable and monetizable.

For those that haven’t had that reckoning, however, the runway is getting shorter by the day. If all of the metrics around how you measure the success of your media business are falling, then it’s a strong indicator you’re in the wrong business. There’s no more time to put off the hard questions about your audience, the value you bring, and how to connect those two in ways that can grow.

In other words, it’s time to panic.

View the full article