Hewlett Packard Enterprise (HPE) Up 18.6% Since Last Earnings Report: Can It Continue?

HPE reported fiscal Q3 2026 non-GAAP EPS of $1.11, up 152.3% year over year and 16.8% above consensus, while revenue rose 33.7% to $12.21 billion. AI demand drove a record $7.6 billion backlog, including $2.4 billion of AI Systems orders and a $3.5 billion post-quarter hyperscaler inferencing contract. HPE raised fiscal 2026 revenue-growth guidance to 34-37%, EPS guidance to $3.75-$3.85, and free-cash-flow outlook to at least $3.75 billion; shares have gained 18.6% since the earnings release.

Analysis

HPE is shifting from a cyclical enterprise-hardware valuation framework toward a hybrid AI infrastructure/networking multiple, but the quality of that rerating depends on backlog conversion rather than bookings. The important incremental signal is that networking demand is supply-constrained while procurement commitments are rising: this creates a 1-3 month revenue-release catalyst if component availability improves, but it also raises working-capital and execution risk if customers defer delivery or suppliers are pre-paid ahead of shipments.

The market is likely underestimating the strategic value of HPE's AI-networking position versus server-only competitors such as SMCI and DELL. As inference clusters proliferate, networking content per deployment rises and customer switching costs increase after architecture selection; this favors HPE's combined compute, routing and campus footprint over the next 6-18 months. ORCL is a demand read-through, but its benefit is more indirect: HPE's infrastructure wins validate Oracle cloud expansion while also increasing Oracle's dependence on timely third-party hardware fulfillment.

The near-term contrarian issue is margin quality. Higher-margin traditional server pricing and favorable networking mix appear to be subsidizing a lower-margin AI-systems ramp, so consensus may be extrapolating peak gross margin rather than management's implied normalization. A post-results pullback is plausible even with continued revenue upside if next-quarter gross margin declines materially, AI backlog does not convert into cash flow, or the large hyperscaler order carries customer-concentration and financing terms that weaken free-cash-flow conversion.

Key Decisions for Investors

• Accumulate HPE on a 5-8% post-earnings or supply-chain-driven pullback; target a 6-12 month holding period. The catalyst is conversion of networking and AI commitments into revenue and free cash flow, with upside from a sustained infrastructure/networking multiple rerating. Exit or reduce if fiscal-2027 operating-margin guidance falls below 14% or free-cash-flow conversion materially misses plan.
• Run a 3-6 month pair trade: long HPE / short SMCI, sized beta-neutral. HPE offers a more diversified enterprise and networking earnings stream plus integration synergies, whereas SMCI remains more exposed to AI-server pricing pressure and component availability. Cover the short leg if SMCI demonstrates durable gross-margin stabilization or HPE reports backlog growth without a corresponding improvement in shipments and cash conversion.
• Use ORCL as a read-through watch rather than a standalone follow-on trade: add to an existing ORCL long only if cloud capex guidance and remaining-performance-obligation growth confirm that infrastructure demand is broadening beyond one project. A slowdown in Oracle cloud bookings or disclosures indicating deployment delays would be an early warning that HPE's large-customer backlog is less immediately monetizable.
• Before adding size to HPE, require disclosure on customer concentration, cancellation provisions, receivables growth and purchase-commitment terms at the next earnings release. These data determine whether reported order momentum represents high-quality demand or a working-capital-intensive build that could cap equity upside despite headline growth.