Sandisk: The Market Is Still Underestimating Its New Earnings Power

Sandisk is characterized as a Strong Buy, with NBM contracts improving earnings visibility and supporting structural margin expansion. Its mix shift toward data-center and AI demand is expected to raise value per bit and improve business quality. Management targets non-GAAP gross margins of 80% and operating margins of 75% by FY2028-FY2030, implying a materially higher normalized earnings base.

Analysis

The key underwriting issue is whether SNDK’s purported contract visibility changes NAND from a spot-price earnings stream into a durable mix/price story. That would justify multiple expansion only if enterprise SSD pricing and customer commitments remain resilient through the next NAND down-cycle; a margin target alone is not evidence of either. The market should demand disclosed contract duration, minimum-volume commitments, cancellation terms, and the portion of revenue exposed to data-center SSDs before capitalizing distant profitability at a premium multiple.

AI is a less direct demand driver for NAND than for HBM/DRAM. Higher AI inference and storage intensity can support enterprise SSD demand, but hyperscalers can also qualify Samsung, Kioxia, Solidigm, and SK hynix aggressively, limiting the ability of any one supplier to retain the incremental economics. The more plausible near-term beneficiary is a disciplined industry-wide NAND pricing environment; the greater risk is that capacity additions or weak client-device demand erase ASP gains faster than cost-per-bit declines.

Over the next 1-3 months, the catalyst path is earnings evidence: data-center revenue mix, enterprise SSD gross margin, contract-backed backlog, and management’s willingness to raise near-term—not long-range—profitability assumptions. Over 6-18 months, the thesis is falsified if NAND ASPs fall for two consecutive quarters without offsetting cost reductions, or if gross margin fails to improve despite claimed mix gains. Contrarian view: the stock may already discount an AI-quality transformation while retaining commodity-memory cyclicality, making downside asymmetric if visibility proves less binding than investors assume.

Key Decisions for Investors

• Do not chase SNDK on the guidance narrative alone; establish a 1-3 month watch position only after earnings disclose contract coverage, enterprise SSD mix, and a tangible upward revision to near-term gross-margin or EPS guidance. Exit if management cannot quantify volume/price protection or if NAND pricing weakens sequentially.
• For AI-memory exposure, prefer a relative-value long MU / short SNDK basket over the next two earnings cycles if SNDK materially outperforms without corroborating enterprise-margin data. MU has more direct high-bandwidth-memory sensitivity, while SNDK remains more exposed to a NAND pricing reversal; reassess if SNDK demonstrates sustained margin expansion for two quarters.
• Use SNDK put spreads dated 3-6 months as a hedge against a post-guidance de-rating if implied volatility is below the stock’s post-earnings realized volatility. The hedge is most attractive ahead of the first report where contract economics can be tested; invalidate if disclosed backlog and enterprise SSD margins materially exceed expectations.
• Monitor NAND contract ASPs, supplier capex/capacity commentary, and hyperscaler storage spending monthly. A broad NAND supply-discipline signal would support a tactical long SNDK; renewed capacity expansion by major Asian suppliers would argue for reducing exposure before the next pricing reset.