This is not a typical earnings preview. On July 14, IBM did something it rarely does: it released preliminary second-quarter financial results rather than waiting for its scheduled call, and the board decided the quarter was such a bust that it got ahead of it. The reaction was historic — IBM's stock dropped 25% in a single day, the worst day in the company's history, with the company ending the week with a market capitalization of just under $200 billion. Some reports pegged the single-day decline as the stock's largest one-day drop since 1968, wiping out roughly $69 billion in market capitalization.
So tomorrow's print is less about "what were the numbers" (largely known) and much more about credibility, guidance, and whether this was a one-quarter stumble or a structural shift in enterprise tech spending away from IBM's model.
From CEO Arvind Krishna's July 14 investor letter, preliminary Q2 2026 figures (vs. consensus):
| Metric | Preliminary Result | Street Estimate |
|---|---|---|
| Revenue | $17.2B, up 1% | ~$17.86B |
| Software revenue | up 5% | — |
| Consulting revenue | flat (+1% constant currency) | — |
| Infrastructure revenue | down 7% | — |
| Operating (non-GAAP) EPS | $2.93 | $3.01–$3.02 |
| GAAP EPS | $2.27, down 2% | — |
| Operating gross margin | 59.4%, down 70 bps | — |
| YTD free cash flow | $4.8B | — |
The magnitude of the miss — particularly in Infrastructure — was the shock. Infrastructure revenue fell 7% as fewer companies bought mainframe hardware and high-margin software for banking and credit-card payment operations, a sharp reversal from the 12% Infrastructure growth (48% Z growth) reported just one quarter earlier.
Management's stated cause is specific and important to interrogate tomorrow. Krishna wrote that in the last few weeks of June, clients shifted their quarterly capex spend toward servers, storage, and memory purchases to secure supply-constrained infrastructure ahead of expected price increases, and admitted the company did not anticipate the magnitude of the capex reprioritization. He also cited clients being distracted with rapidly-evolving, industry-wide cybersecurity concerns. Beyond the macro excuse, he owned execution: "we did not adapt and move quickly enough, and numerous large deals failed to close on the timelines we expected, driving the majority of our shortfall."
Krishna did highlight pockets of strength buried in the release: Red Hat revenue growth accelerated sequentially to 11%, HashiCorp and Confluent delivered strong performance, Distributed Infrastructure had its best performance in reported history (up 37%, with a ~$500 million exit backlog), and z17 remained at nearly 130% program-to-program, well ahead of the record-setting z16 cycle.
The debate investors will want resolved on the call is whether IBM is a victim of an industry-wide AI capex squeeze or facing company-specific execution/relevance problems. Evidence points both ways:
IBM shares are a study in whiplash: from roughly $320 in early June, the stock slid steadily into the high-$260s/low-$270s by mid-June (independent of the eventual warning), rallied back above $300 by July 8–10 on momentum and an analyst initiation, then cratered from $290 to $217 in a single session on July 14 and has drifted lower since, closing around $210.54 on July 21. Notably, just before the warning, shares had rallied 25% in the second quarter, bringing them near record highs, though the stock was down less than 1% year-to-date, trailing the S&P 500's 10%+ gain — meaning a lot of the "AI infrastructure winner" premium had already been unwound heading into the print. As Mizuho's Dan O'Regan noted, while the stock had become a crowded AI infrastructure winner trading near all-time highs making it vulnerable to punishment for execution issues, the magnitude of the decline suggests the market is pricing in a much more prolonged slowdown than management has implied — a gap tomorrow's call needs to close.
The actual EPS/revenue print is stale news. The report that matters tomorrow is about narrative repair: can Krishna and CFO Jim Kavanaugh convincingly separate a temporary, industry-wide AI-capex-driven pull-forward from a structural erosion of IBM's software and mainframe annuity model? Watch closely for (1) explicit full-year guidance (raised, maintained, or cut), (2) any quantification of how much of the "shifted" Q2 demand has already come back in July, (3) tone and specificity on Software durability versus AI-disintermediation fears, and (4) any signals on capital return, M&A, or portfolio actions that address the growing activist/breakup chatter. Given the stock's historic drop already priced in a severe outcome, the bar for a "relief rally" may be lower than usual — but so is management's margin for another disappointment.