Down 23%, Should You Buy the Dip on Sandisk Stock?

Sandisk shares are down 23% from their June 2026 high despite fiscal Q4 revenue rising 372% year over year and adjusted EPS increasing 68% sequentially to $39.25. Data-center demand grew from 12% to 38% of bit volume by fiscal 2026, while eight multiyear customer agreements secure nearly $94 billion of future revenue and could cover roughly two-thirds of bit volume by fiscal 2028. Management targets mid-to-high-teens revenue growth from fiscal 2028 to 2030 and approximately 50% free-cash-flow margins, though uncontracted NAND volumes remain exposed to pricing and capacity-cycle risks.

Analysis

The investable question is whether SNDK has transitioned from a spot-NAND earnings lever into a contracted enterprise-storage compounder. That warrants a multiple re-rating only if the disclosed commitments contain enforceable minimum purchase volumes, pricing floors, and limited customer termination rights; headline contract value alone can materially overstate backlog quality. The next 1-3 months should focus on customer concentration, contract accounting, deferred revenue, and whether enterprise gross-margin expansion persists as mix shifts.

A tighter NAND supply environment benefits SNDK disproportionately if it has already committed capacity at favorable economics, but it creates a second-order risk: rivals Samsung Electronics, SK hynix and Kioxia may respond with aggressive capacity additions once pricing and utilization improve. NAND cycles typically turn on supply discipline rather than demand narratives; a meaningful increase in industry wafer-capacity guidance or SNDK capex would undermine the claimed durability of cash generation over the following 6-18 months.

The recent decline is not automatically attractive after an extreme prior rerating: forward P/E is a weak anchor when earnings sit near a memory-cycle peak. Consensus is likely underestimating enterprise mix durability but also underweighting the risk that locked-in volumes cap upside participation if NAND pricing spikes further. The thesis is falsified by two consecutive quarters of enterprise-margin compression, weaker contracted-volume mix, or fiscal-2027 capex materially above depreciation without corresponding customer prepayments.

Key Decisions for Investors

• Establish only a starter long SNDK position after the next earnings release, sized at 50% of normal semiconductor risk, contingent on disclosure of take-or-pay terms and stable enterprise gross margin. Target 20-30% upside over 6-12 months if the market accepts a structurally higher earnings multiple; exit if management cuts contracted-volume expectations or guides material capacity expansion.
• Use a relative-value expression: long SNDK / short SOXX in equal beta-adjusted dollars for 3-6 months. This isolates company-specific contract and margin validation from broad AI-semiconductor valuation risk; close if NAND pricing weakens while DRAM/HBM-led SOXX strength persists, indicating the storage thesis is decoupling negatively.
• Do not add on a spot-NAND price spike alone. Add only if quarterly free-cash-flow conversion confirms that working-capital needs and capex are consistent with the long-run margin framework; this is the key missing verification behind the bullish valuation case.
• Set an industry alert for new NAND fab or major utilization-restoration announcements from Samsung, SK hynix, Kioxia, or Micron. A coordinated supply response is a 6-18 month catalyst to reduce SNDK exposure or initiate a tactical hedge through short semiconductor-equipment exposure only after capex orders become observable.