Sandisk Stock Is Still Down 50% From Its Highs. Is the Memory Winner Still a Screaming Buy Before September?

The article argues Sandisk (SNDK) is benefiting from ongoing NAND/SSD supply constraints tied to AI data-center demand, with memory pricing remaining elevated into the post-2027 period. It cites Amazon raising 2026 capex from $200B to $220B due to higher memory chip prices and claims Wall Street expects SNDK revenue growth of 151% in fiscal 2027 (ending June 30). Despite noting the market’s fear of a later chip-demand slowdown, the piece frames the current 50% drawdown from highs as a buy opportunity and expects a stock rebound after the pricing imbalance persists.

Analysis

The tradeable mechanism is not “AI demand” in the abstract; it’s a temporary re-rating of a commoditized input where tight supply lets the marginal supplier capture outsized price. That makes SNDK the direct winner, but the second-order losers are hyperscalers and storage-heavy hardware buyers: rising NAND spend can quietly pressure cloud gross margins and force procurement normalization later, which matters more for AMZN than for the broader AI narrative. Peers like MU and WDC should also participate, but SNDK may have the cleanest operating leverage if its mix is more exposed to SSD pricing.

The key risk is cycle timing. Memory upcycles usually end not when demand disappears, but when supply response hits with a lag; that is a 6-12 month problem, not necessarily a 2027 problem. The near-term catalyst path is next-quarter ASP and gross margin commentary: if pricing is still inflecting up, the stock can keep rerating; if management starts talking about normalization or customers pushing out orders, the multiple can compress fast even if revenue stays elevated.

Consensus may be underestimating reflexivity: higher memory prices can pull forward buying and make the shortage look more durable than it is. That supports the move over the next 1-3 months, but it also makes SNDK vulnerable to a sharp air-pocket once procurement teams finish restocking. The falsifier is simple: any sign of capacity additions, weaker SSD pricing, or margin guide deceleration would argue this is a trade, not an investment.

Key Decisions for Investors

• Long SNDK on pullbacks for a 1-3 month trade into the next earnings/guidance cycle; use common stock if you want linear upside, or a call spread if you want to cap cycle-reversal risk.
• Add a small relative-value hedge: long SNDK / short AMZN over 1-2 quarters to isolate memory inflation pressure on hyperscaler margins; cut the hedge if AWS margin commentary remains resilient.
• Use MU and WDC as confirmation signals rather than chasing them outright; if either company signals tightening ASPs and no capacity relief, the SNDK thesis is intact.
• Set a hard watch item on gross margin and ASP guide: any deceleration in those metrics is the main thesis breaker and should trigger profit-taking immediately.