Sandisk: Continued Memory Boom On Data Center Buildout

Sandisk’s revenue growth is being supported by AI-driven NAND memory shortages that are expected to persist until at least 2027. The company’s Flash Venture expansion with Kioxia helps secure NAND supply and adds capacity plans to double wafer output by FY2029, strengthening supply-chain resilience. Sandisk also plans a 332-layer 10th Gen NAND ramp alongside high-bandwidth flash products to improve competitiveness as industry conditions shift.

Analysis

Near term, the setup is more about pricing discipline than unit growth. If AI-related demand is genuinely soaking up enterprise flash, SNDK’s main benefit is not higher bit shipments but a better mix and less frequent forced discounting, which can lift gross margin faster than revenue. The JV with Kioxia matters because shared capex reduces the chance that SNDK has to fund a capacity race alone; that lowers downside in a cyclical downturn and improves cash conversion versus standalone peers.

The second-order winner is the NAND equipment and specialty materials chain over a 6-18 month horizon if the wafer-capacity expansion is real and not just aspirational. LRCX, AMAT, and KLAC can benefit from a longer spending runway, while legacy storage OEMs and consumer-device makers face a tougher procurement environment if supply stays tight. But the same capacity build is a future supply overhang: by FY2029, if AI demand normalizes or shifts toward DRAM/HBM instead of NAND, today’s scarcity premium could flip into price pressure.

The market may be overconfident that every AI memory shortage is durable and NAND-specific. The cleaner contrarian is that this is a good stock story but a mediocre cyclical one: the stock can rerate on tighter supply, yet the industry still has a history of overshooting into excess capacity. What would falsify the bullish case is any sign that NAND ASPs soften before the next guidance cycle, or that 332-layer ramp/yield issues offset the supply benefit. On a 1-3 month horizon, the trade is mainly about whether management can translate the narrative into margin revision, not just revenue language.

Key Decisions for Investors

• Long SNDK on pullbacks over the next 2-6 weeks; prefer a staged entry ahead of the next earnings/guidance event. Upside comes from gross-margin revision more than top-line surprise; thesis fails if NAND ASPs roll over or management frames the shortage as temporary.
• Use a call spread rather than outright stock if implied volatility is rich: 3-6 month SNDK call spread to capture a gradual rerating while limiting downside if the AI-memory narrative is overextended.
• Watch AMAT/LRCX/KLAC as secondary beneficiaries over the next 6-18 months; if capex commentary from SNDK/Kioxia firms up, these names can outperform on a longer equipment cycle before bit supply catches up.
• Avoid chasing a broad long NAND basket right now; the cleaner expression is single-name SNDK versus no hedge, because the future FY2029 capacity build could become an industry-wide overhang and compress returns for the entire complex.