Meta’s Identity Crisis Revealed by Traffic Data

On Friday, May 8, 2026, The New York Times published a guest essay by investigative journalist Julia Angwin titled “Meta Is Dying.” She points out that Meta lost daily active users in Q1 2026, dropping from 3.58 billion in Q4 2025 to 3.56 billion. Angwin views this as the start of a prolonged, gradual decline, likening the company’s path to AOL in 2003 and Yahoo in 2015: still technically alive and profitable, but entering what she calls the “zombie era.”

She could be correct. And if so, Theodore Levitt explained why this would happen 66 years ago.

The Lesson Meta Never Learned

In 1960, Harvard Business School professor Theodore Levitt published “Marketing Myopia” in the Harvard Business Review. His core argument was that companies fail not because demand vanishes, but because they define their business too narrowly. Railroads collapsed because they saw themselves in the railroad business, not the transportation business. Trolley car companies were overtaken by automobiles they could have pioneered. “People don’t want a quarter-inch drill,” Levitt wrote. “They want a quarter-inch hole.”

Now examine Meta’s six major pivots over 22 years and ask: What business did Mark Zuckerberg actually think he was in?

In 2021, he declared it was “the metaverse business.” That bet, through Reality Labs, has since accumulated roughly $80 billion in operating losses. Users didn’t buy in. In 2023, he pivoted to generative AI, committing over $100 billion to building models that, as Angwin notes, currently underperform competitors. Q1 2026 results show record revenue of $56.3 billion, up 33% year over year, but also $33.44 billion in total costs, a 35% increase, and an AI spending outlook that has unsettled investors.

The revenue appears strong. The trajectory, however, shows a company that keeps pivoting to new product definitions while its core users quietly disengage.

What the Traffic Data Actually Shows

This is where opinion meets evidence, and the Similarweb traffic data for March 2026 is telling.

Google leads globally with 86.9 billion monthly visits. YouTube follows with 29.3 billion. Facebook comes in third at 11.9 billion, and Instagram is fourth at 7.1 billion. That gap between Google and Facebook is the data equivalent of what Levitt described. Google defined itself as being in the information access business. Facebook defined itself as being in the social network business. One definition scales indefinitely. The other runs out of room.

The AI category data is even more revealing. ChatGPT records 5.7 billion monthly visits globally, with year-over-year growth of 28.5%. Gemini is growing sharply at 283.8% YoY. Claude.ai jumped 423.7% to 613.7 million visits YoY.

Meta.ai does not appear in the top 100 most-visited websites.

Meta spent $100 billion entering the AI race. It is not winning it.

The Squeeze Play Angwin Describes

When an aging platform’s user base starts to shrink, the typical response is to monetize harder. Angwin documents this clearly. Meta’s Q1 ad impressions increased 19% year over year while average ad prices rose 12%. Revenue per user jumped 27%. The company is cramming more ads onto its platforms and charging advertisers more for each one.

This move maximizes short-term revenue while accelerating long-term decline. More ads mean a worse user experience. A worse experience means slower growth. Slower growth means the ad inventory eventually stops expanding. Levitt described this as the trap companies fall into when they focus on selling their current product harder instead of understanding what customers actually need.

For digital marketers and SEO professionals, this creates a near-term concern. Meta’s Advantage+ advertising suite delivers genuinely strong performance data: a $4.52 return per dollar spent, 22% higher than comparable manual campaigns, according to Meta’s own earnings reports. But those returns depend on a healthy, engaged user base generating meaningful behavioral signals. If the user base contracts and ad load increases simultaneously, signal quality degrades, and performance follows.

The Counterargument Worth Taking Seriously

Angwin’s essay is persuasive, but she is writing opinion, not analysis, and the full Q1 picture is more complex than “dying” suggests. Year over year, Meta’s daily active user base still grew 4%. The quarter-over-quarter decline has a partially verifiable explanation in internet disruptions in Iran and Russia’s WhatsApp ban. Revenue growth of 33% is not the profile of a company in terminal decline.

What it is, is the profile of a company spending at a scale that requires the growth to continue, while its AI investments have not yet produced meaningful new revenue streams. As the Wall Street Journal’s Asa Fitch observed this week, “the spending growth looks increasingly unsustainable.”

Levitt’s lesson wasn’t that myopic companies always die quickly. AOL and Yahoo lingered for years. The lesson was that once a company loses the plot on what business it’s actually in, recovery becomes structurally difficult. Every dollar spent defending the wrong definition is a dollar not spent understanding the customer.

The question Levitt would ask isn’t whether Meta is dying. It’s whether Meta has ever clearly understood what business it was actually in. Across six pivots in 22 years, the answer appears to be: not consistently.

That uncertainty is now visible in the traffic data. And traffic data doesn’t lie.