Zoom Didn't Crash After COVID. It Split in Two — and Only Half Is Dying.

On October 19, 2020, a single share of Zoom closed at $568.34 — the high-water mark of the pandemic's defining stock.⁵ By mid-2026 that same share traded near $83, a fall of roughly 85%.⁵ Every obituary writes itself from there: Zoom was a moment, the moment passed, the company is fading with the masks. It's a clean story. It's also wrong about the one thing that matters — because while the stock was losing five-sixths of its value, the business was setting a new revenue record every single year.

The popular version says Zoom revenue fell after COVID and that Microsoft Teams took its crown. Neither happened. Revenue went $4.10B, $4.39B, $4.53B, $4.67B — up, up, up, every fiscal year a new high.¹² What collapsed was the growth rate and the share price, not the income statement. The real story isn't a death. It's a split.

One company, two opposite weather systems

Zoom no longer behaves like one business. It behaves like two stapled together, pointed in opposite directions. The Enterprise segment — the seats sold to companies, with contracts and procurement and IT — grew about 5% in fiscal 2025, to roughly $2.75 billion. The Online segment — the consumers and small accounts who signed up with a credit card during lockdown — was essentially flat for the year and went slightly negative in the fourth quarter.³ Rewind to fiscal 2023 and the divergence is starker still: Enterprise revenue grew 24% while the Online business was already shrinking, down 10% year over year in that quarter.⁴ Same logo. Opposite math.

Here is the thesis, in one line a smart friend could repeat at dinner: Zoom isn't dying of a COVID hangover — it's bifurcating, and its survival depends on whether the growing enterprise half can outrun the bleeding consumer half before Microsoft's bundling closes the door. The aggregate number — 3.1% total growth in fiscal 2025³ — is a fiction of averaging, the comfortable temperature you get from one foot in ice and one in fire. Investors who watch only the blended figure are reading the thermostat and missing the storm.

Why the consumer half was always going to leak

The mechanism behind the Online decline isn't mismanagement; it's gravity. A consumer who subscribed to host a 2020 family quiz night or a side-hustle webinar has no contract, no IT department, and no switching cost. They churn the moment the meeting they signed up for stops mattering. Zoom's own numbers show how stubborn this is: even after years of cleanup, Online monthly churn was still 2.8% in the first quarter of fiscal 2026 — improved, but compounding.⁶ A 2.8% monthly leak quietly empties a third of the cohort over a year. That is the structural truth of a consumer tier built during an emergency: it was rented attention, and the emergency was the lease.

So the Online segment isn't a problem to be fixed. It's a windfall that is reverting to its natural level, and no amount of feature work pulls a lockdown back into existence. Which is why the only question that matters is happening on the other side of the company — at the enterprise.

The race against a bundle, not a better mousetrap

At the enterprise, Zoom's bet is to stop being a video-call app and become a platform — Zoom Phone, Contact Center, the Workvivo employee-communications layer, and AI Companion stitched across all of it. The traction is real if still small: AI Companion's monthly active users grew 68% quarter over quarter in the fourth quarter of fiscal 2025, by the company's own account.³ Operating income tells the same story of a maturing business learning to make money: income from operations climbed from $245M in fiscal 2023 to $525M in fiscal 2024 to $813M in fiscal 2025, with operating cash flow up nearly 22% to $1.95 billion.²³ This is not a company running out of road. It's a company building a second engine while the first one cools.

The threat isn't that Zoom makes a worse product. It's that Microsoft makes Teams effectively free. The competitive weapon is bundling: Teams arrives pre-installed inside Microsoft 365, the suite a corporation has already bought. That is why Teams' market share leapt about 31% in a single year to roughly 32%, while Zoom slipped a couple of points to around 56%.⁸ Zoom still leads. But it leads against an opponent whose marginal cost of acquiring the next enterprise seat is zero, because the customer already signed the check. You cannot out-feature a line item that's already paid for.

The honest counter: maybe the leader is just slowly losing

The fair objection is that 'bifurcation' is a flattering word for 'slow erosion with extra steps.' Zoom's share fell while Teams' rose; the enterprise half grew at a mid-single-digit pace that a bundler can simply outwait; and Zoom just announced it will stop reporting Enterprise customer count as a KPI starting in fiscal 2026⁶ — exactly the kind of disclosure a company retires when the trend embarrasses it. All true, and worth sitting with. But the counter has a tell. A company being routed doesn't grow operating income from $245M to $813M in two years² or guide to another revenue record near $4.8 billion.⁶ The pessimist's case requires the enterprise engine to stall before it scales — and the data so far shows it scaling. The optimist's case requires Zoom to keep winning net-new platform seats faster than the consumer tier bleeds. Both are live. That's what makes this a genuine fork rather than a foregone conclusion.

Zoom's stock crashed because the market had priced a verb — 'to zoom' — as if a moment could be owned forever. The business underneath did something far less dramatic and far more interesting: it shed the customers it was never going to keep and went hunting for the ones it could. The consumer half is doing what windfalls do. The enterprise half is doing what survivors do. Whether Zoom lives is not a question about COVID anymore. It's a question about a race — between a platform learning to charge for itself and a bundle that never has to. The verb is gone. What's left is the arithmetic, and the arithmetic is still being written.