Zoom's AI Pivot Is Real. The Revenue Problem Is Still the Hurdle

Zoom's second-quarter report was a beat against a modest hurdle, and the market mostly shrugged: revenue of $1.28 billion, up 4.9% year over year, non-GAAP EPS of $1.55, about 5% above the consensus of $1.48. Management raised full-year guidance, put another $1.3 billion on the buyback pile, and the stock barely moved. That quiet is itself the tell. After the pandemic multiple collapsed, the market stopped asking whether Zoom could keep its users. The new question is whether this company can ever convert meetings into money again.

The answer coming out of Q2 is yes—just not yet at the scale the stock is priced for.

Where the growth actually lives

Peel the 4.9% apart and the story is inside the enterprise line. Enterprise revenue rose 7.8% to $787.5 million, the fastest rate in three years, while the smaller online/consumer business grew just 0.6%. That split matters because the enterprise segment is where Zoom's AI bet is being sold, and it is now roughly three-fifths of total revenue.

The driver is not the free AI Companion that summarizes your meetings. That product's licensed monthly active users grew 125% year over year, and it tripled during the prior fiscal year—but it is a retention tool, bundled into existing plans, not a meter. The paid engine sits in Zoom's customer-experience (CX) and contact-center business, where companies deploy virtual agents and automated workflows that actually complete work instead of summarizing it. The evidence of traction is concrete: customer count for Zoom's Virtual Agent rose 256% year over year, paid AI was included in each of its top ten CX deals, and the CX portfolio is booking high-double-digit annual recurring revenue growth.

The deeper question is whether it can move the headline

Here is the tension a reader should feel before paying for the AI narrative. Net dollar expansion—how much revenue Zoom collects from an existing enterprise customer a year later—was 99%, below the 100% bar. In plain terms, existing customers are still shrinking slightly on a net basis, which means the enterprise acceleration is coming mostly from new and expanded logos, not from a growing base spending more. That is a legitimate early-stage sign, but it is also the reason total revenue is stuck at mid-single digits.

And the AI monetization, while real, is still young. Zoom built its franchise giving away the flagship AI feature; the money has to come from selling agents and contact-center seats on top. Management has been explicit that the contact center offers "the most immediate pulls" of incremental AI monetization, with deals closing now rather than waiting in pipeline. That is the right sequencing—monetize workflow, not summarization—but it means growth will be lumpy and deal-dependent until it isn't.

What the price is asking for

Strip the accounting noise and Zoom is a high-margin cash machine wearing a growth-stock price. Full-year guidance points to roughly $5.09 billion of revenue, non-GAAP EPS of $6.08–$6.12, and free cash flow of about $1.8 billion. Against an ~$28 billion market cap, that is a mid-teens earnings multiple and a free-cash-flow yield in the mid-single digits—nothing crazy for a profitable, buyback-heavy company, but also not a bargain for one growing 5%.

The multiple is the market's bet on a rerating: that the CX and agentic-AI expansion eventually lifts total growth toward the high single digits and, crucially, pushes net dollar expansion back above 100. Nothing in Q2 proved that will happen. The quarter offered evidence the direction is right, not proof the magnitude will land.

The checkpoint that breaks the narrative

So the investor's "so what" is narrow. Watch whether enterprise revenue keeps accelerating quarter over quarter and whether net dollar expansion creeps above 100%—the two numbers that would turn a three-year-best 7.8% into a structural re-acceleration instead of a one-quarter popsicle. On the risk side, the same figure that protects the story is the one that breaks it: if the CX deals keep growing double digits but total revenue stays parked at ~5%, you will know the AI engine is real but too small to matter, and the stock will keep trading as a mature company, not an AI one.

Zoom has answered the "Will anyone pay for this?" question. The next quarter answers "Will it be enough?"