Rackspace Technology heads into its third-quarter report with a story that has rarely looked more ambitious and a stock that has rarely looked less trusted. Last quarter, management turned a tentative AMD memorandum into a definitive 30-megawatt GPU deployment plan and attached real economics to it for the first time. The market answered by cutting the share price roughly in half. The November 9 report, due after the close, is the first chance to show whether the AI pivot is building real momentum or simply papering over a shrinking core business.
The near-term numbers have been moving the wrong way. Full-year revenue guidance has slid from $2.6 billion to $2.7 billion in February to $2.45 billion to $2.55 billion, which implies roughly a 7% decline. Adjusted EBITDA, non-GAAP operating profit and free cash flow targets have all been trimmed along the way, with the cash flow outlook nearly halved to $50 million to $70 million. Management now expects a full-year loss of $0.25 to $0.30 per share. The Street's consensus loss of $0.18 is noticeably more forgiving than the company's own range. That gap matters: either analysts expect conservatism to give way to an upside surprise, or estimates still need to come down. For context, the year-ago quarter produced a loss of $0.05 per share on revenue of $671.2 million, and the second quarter just reported delivered a loss of $0.08 on $670.1 million.
The quarter's toplines carry a built-in complication. Private cloud beat its second-quarter guide handily, but management said the beat came from a one-time embedded-lease accounting item that pulled revenue forward from the second half. As a result, a softer private cloud number this time would not necessarily signal deterioration, but it will make the segment harder to read. Public cloud is being deliberately shrunk, with a guide calling for about a 13% decline as low-margin resale work is exited. The cleaner test is profitability. Consolidated non-GAAP gross margin fell to 18.6% last quarter and private cloud gross margin dropped 460 basis points year over year. If exiting low-margin revenue is the strategy, margins should start to show the benefit. Further compression would undercut the core rationale for accepting a smaller company.
On the growth side, the evidence needed is concrete. Management said hardware for the first roughly 2 megawatts had already been ordered, with about $75 million of capex aimed at a year-end deployment. Any update on delivery timing, data center readiness and, most importantly, signed customers to consume that capacity will carry weight. The company has pitched $15 million to $20 million in annual revenue per megawatt at EBITDA margins above 50%, but those figures remain illustrative until contracts land. Progress in the Palantir pipeline, early RackAI adoption, and the continued ramp of the expanded AdventHealth migration would help show that enterprise demand is real rather than aspirational.
Financing is the shadow over all of it. Leverage stood at 8.3 times earlier this year, and the AI buildout leans on equipment and asset-backed facilities layered on top of that. Investors will want clarity on cash burn, liquidity and how much additional debt the 15-megawatt target for 2027 will require.
Sentiment has collapsed from strongly bullish heading into the last report to essentially neutral now, and the stock at $2.62 sits well below its 200-day average of $3.18 and just above its post-earnings low of $2.55. Expectations are depressed, which lowers the bar, but credibility is thin. The report will hinge on whether Rackspace can hold its reduced guidance, stabilize margins and show signed demand for its AMD capacity. Deliver that, and the AI narrative earns a hearing. Cut the outlook again, and the market is likely to treat the pivot as a costly bet on a balance sheet with little room for error.