Sandisk's NAND Cycle Just Changed Its Rules

Sandisk reported Q4 revenue up 372% to $8.97B alongside adjusted free cash flow of $5.04B. Data center revenue rose 437% in fiscal 2026 as AI infrastructure shifted the NAND demand mix, with new business models covering >50% of fiscal 2027 bits and ~two-thirds of fiscal 2028 supply. Overall, the combination of rapid revenue growth and very strong cash generation points to a materially improved demand and monetization outlook.

Analysis

The key market change is not the print itself; it is the mix shift toward data-center NAND, which should make SNDK look less like a commodity memory trader and more like a leveraged AI infrastructure supplier. That tends to expand valuation multiples if the market believes the revenue base is becoming contract-backed and less exposed to consumer inventory cycles. The second-order winner is the NAND ecosystem generally, but SNDK has the clearest operating leverage because every point of pricing and utilization improvement should drop quickly into cash flow.

The setup is still cyclical. A lot of the apparent strength could be peak-cycle working capital release plus a favorable mix, so the first real test is whether data-center demand stays above consumer/auto weakness for the next 1-3 quarters. If peer capex or wafer starts rise meaningfully, NAND pricing can turn faster than the market expects, and the current rerating thesis would become vulnerable within 6-12 months. Watch for any commentary on ASPs, bit growth, or contract pricing that suggests visibility is being bought by capping upside.

The contrarian view is that consensus may be underestimating how much of the upside is already in the tape after a move this large, while overestimating the durability of “new business model” coverage if it simply locks in current favorable terms rather than creating incremental margin. The cleaner expression is relative outperformance versus the broader semi complex, not a blind chase of the absolute move. If the market starts treating SNDK as a structural AI winner, that would also pressure adjacent enterprise storage vendors and any customer with NAND-heavy BOM exposure.

Key Decisions for Investors

• Buy SNDK on a 1-2 day post-earnings pullback, targeting a 15-20% upside over 1-3 months if data-center mix remains the dominant growth engine; stop out if management guides to slower enterprise demand or softer ASPs.
• For beta control, express the view as long SNDK / short SOXX over the next quarter; this isolates idiosyncratic NAND outperformance if the market keeps rewarding AI-storage exposure.
• Do not chase with cash equity after a vertical move; if entering here, prefer 3-6 month call spreads to cap downside while keeping convexity to another leg up on contract wins or margin expansion.
• Set a falsifier alert: if next quarter shows sequential deceleration in data-center revenue or any evidence of NAND pricing normalization, reduce exposure immediately because the trade becomes a cyclical fade rather than a structural rerate.