Up Nearly 700% This Year, Is Sandisk Still Worth Buying?

Sandisk is benefiting from an AI-driven memory chip shortage, with demand for SSDs and NAND chips outstripping supply and pushing prices higher. The article says Wall Street expects EPS of $177.84 by fiscal 2027, which could imply a share price above $3,500 if the stock trades at 20x earnings, roughly doubling from current levels. The outlook remains dependent on the supply bottleneck lasting into 2028; if supply normalizes and prices fall, upside could narrow.

Analysis

The market is pricing SNDK as if the current tightness in NAND is a multi-quarter structural regime, not a cyclical squeeze. That can work until it doesn’t: semicap and memory pricing tend to inflect faster than consensus expects once capacity additions and customer inventory normalization collide, so the key risk is not demand collapse but margin compression from even modest supply recovery. In other words, the setup is less about whether AI storage demand exists and more about how quickly incremental wafer starts and channel destocking can flatten the pricing curve.

The second-order winner is not just SNDK; it’s the entire AI storage stack, including SSD-adjacent component suppliers and data-center OEMs that need to secure inventory early. But the margin expansion here may be self-limiting because customers will eventually redesign around cost, shifting mix toward higher-density, more efficient architectures or delaying non-critical builds. If hyperscaler capex growth slows from “accelerating” to merely “high,” NAND pricing can still remain elevated, but the multiple will likely de-rate before earnings peak because the market discounts normalized conditions 6-12 months ahead.

The consensus seems to underappreciate how much of the upside is already embedded in an earnings power story that assumes scarcity lasts into 2028. That makes the stock more vulnerable to a supply-side surprise than to a demand miss. The clean contrarian take is that the easy money has likely been made; the next leg higher requires not just strong orders, but proof that the shortage survives the industry’s natural instinct to overbuild into strength.

Relative value is more interesting than outright direction: NVDA and INTC are marginal beneficiaries of the same AI spend cycle, but SNDK has the most direct leverage to storage pricing and therefore the highest beta to a reversal in component inflation. For investors who want exposure to AI infrastructure without paying peak scarcity multiples, a basket or pair structure is preferable to a full-size single-name long.

Key Decisions for Investors

• Avoid initiating a new outright long in SNDK at current levels; if held, use a 3-6 month trailing stop or trim into strength because the stock is most exposed to any NAND price inflection.
• For tactical upside, buy SNDK call spreads 6-9 months out rather than stock; the thesis works if scarcity persists, but options cap downside if pricing normalizes faster than expected.
• Pair trade: long NVDA / short SNDK into the next 1-2 quarters if you want AI exposure with better earnings visibility; NVDA benefits from secular capex, while SNDK is closest to a cycle peak.
• Watch for any indication of NAND capacity expansion or customer inventory rebuild completion over the next 90-180 days; that would be the signal to reduce SNDK exposure aggressively.
• For higher-conviction exposure to the storage shortage, consider a basket long in AI infrastructure beneficiaries rather than concentrated SNDK, to dilute single-name commodity risk.