Dell's ISG is the primary growth engine and central investment thesis. As NVIDIA's preferred OEM partner, Dell ships more enterprise AI servers than any other vendor. The PowerEdge platform with NVIDIA Blackwell GPUs is the standard enterprise AI infrastructure choice. Storage platforms benefit from AI data pipeline requirements.
ODM bypass and margin compression are the twin ISG risks
If hyperscalers continue shifting to direct ODM procurement for AI servers, Dell loses the highest-volume AI server segment. Meanwhile, even in enterprise AI where Dell leads, NVIDIA GPU BOM dominance leaves thin margins for the OEM assembler. The value accrues to NVIDIA, not Dell.
What is Dell's actual AI server gross margin vs traditional server margin?
Dell's AI server business is built on the NVIDIA partnership and the PowerEdge platform. Dell is the preferred OEM for enterprise NVIDIA GPU deployments, shipping PowerEdge XE9680 (8x GPU) and other configurations. The backlog exceeds $20B, driven by enterprises deploying on-premise AI infrastructure. Dell is also qualifying AMD Instinct GPUs as an alternative, reducing NVIDIA dependency.
Dell PowerEdge XE9680 is the industry-standard 8-GPU AI server, available with NVIDIA H100, H200, and B200 GPUs. Also qualified with AMD Instinct MI300X. Starting from ~$300K per u...
Dell's traditional server and storage business generates approximately $25-28B in annual revenue, roughly flat YoY. PowerEdge (x86 servers), PowerStore (block storage), PowerScale (file/NAS), and PowerFlex (software-defined) are the core platforms. This is a mature, competitive market where Dell competes primarily with HPE, Lenovo, and NetApp.
Dell traditional server revenue (non-AI) approximately $20B in FY2026, roughly flat. Dell maintains #1 or #2 position in enterprise x86 servers depending on region. Competition fro...
Dell's ISG faces a structural margin challenge: AI servers carry thin 12-18% gross margins because NVIDIA GPUs represent 60-80% of the bill of materials. Meanwhile, hyperscalers increasingly bypass OEMs like Dell and procure directly from Taiwanese ODMs (Quanta, Foxconn, Wiwynn) at 10-20% lower cost. Dell's value proposition is the full-stack solution (server + storage + networking + services) that enterprises need but hyperscalers don't.
AI server gross margins estimated at 12-18%, compared to 25-30% for traditional servers. The margin compression is structural: NVIDIA GPUs account for 60-80% of AI server BOM cost,...