IBM told investors Tuesday that it expects to fall short of its own quarter, projecting adjusted profit of $2.93 a share on $17.2 billion in revenue against FactSet consensus of $3.01 and $17.86 billion, and the stock fell roughly 23 percent in morning trading [1]. The gap is small in dollars and large in signal: a company that has spent two years selling itself as an AI and quantum turnaround told the market it could not hit its own line.

The preliminary release, rather than a full quarterly report, is the part that matters most and the part being least discussed. IBM has published a top-line miss and a management explanation, but not the segment breakdown, cash flow, or forward guidance that would show whether the shortfall sits in software, consulting, or infrastructure — the difference between a timing problem and a demand problem [1].

According to AP, chief executive Arvind Krishna attributed the miss to deals that slipped late in the quarter and to shifts in how clients are spending, saying customers moved money toward servers, storage, and memory ahead of price increases rather than toward the higher-margin software and consulting IBM leans on [1]. In that telling the revenue is deferred, not lost — orders expected to close simply did not book inside the window. Orders expected to close are not booked revenue, and a preliminary number cannot yet show how much of the pipeline survived into July.

That distinction is precisely what the day's trading erased. On X, the 23 percent drop is being read as a referendum on the AI-and-quantum narrative — proof, in the market's shorthand, that IBM's reinvention is stalling — while the company's own account frames the same numbers as a spending-mix quirk driven by clients front-running price hikes. The selloff prices the harsher story; the disclosure supports the milder one; neither can be settled until the full report arrives.

The stakes run past one session. A 23 percent move on a preliminary figure resets expectations for the entire second-half guidance IBM has not yet given, and it lands on a company whose valuation increasingly depends on convincing investors that consulting and software growth can outrun a slowing hardware cycle. If the deferred deals close in the current quarter, Tuesday looks like an overreaction; if they were demand that evaporated, the market got there first.

For now the record shows a self-reported miss, a CEO's explanation, and a morning share move. The segment detail, the cash flow, and the guidance that would tell investors which story is true remain unreleased [1].

-- THEO KAPLAN, San Francisco