Is Sandisk Stock a Buy on Surging Revenue and New Contract Structures?

Sandisk reported fiscal Q3 revenue of $6.0 billion, up 251% year over year, with gross margin expanding to 78.4% from 22.5% and adjusted EPS swinging to $23.41 from a $0.30 loss. Q4 guidance calls for revenue of $7.75 billion to $8.25 billion and adjusted EPS of $30 to $33, while the company signed five long-term NBM contracts totaling at least $42 billion in backlog and covering over one-third of next year's BiCS production. The agreements should improve visibility and reduce cyclicality as NAND pricing remains in a supercycle.

Analysis

SNDK’s setup is less about a one-quarter earnings beat and more about a structural reset in pricing power. The long-term contracts matter because they convert what was a pure spot-market exposure into a blended book with visible volume, floor pricing, and counterparty support; that should compress the market’s implied cyclicality discount over the next 2-4 quarters. In other words, the multiple can rerate even if NAND prices merely stay elevated rather than continue climbing.

The second-order winner is the broader AI storage stack: if NAND supply remains tight and BiCS allocation gets locked in, enterprise SSD OEMs and hyperscalers will be forced to prioritize design-ins and inventory planning earlier than usual. That tends to benefit vertically integrated platform vendors and the strongest controller/SSD ecosystem players, while squeezing smaller module assemblers that lack allocation certainty and balance-sheet firepower. For semiconductor peers, the real read-through is that capex discipline across memory is now being rewarded, which supports a multi-quarter pricing umbrella.

The main risk is that consensus may be underestimating how quickly this turns from supercycle to normalization once the big producers re-accelerate NAND output or if AI capex digestion slows in 2H26. Because the stock has already repriced sharply, the near-term asymmetry is less about operational upside and more about any sign that contract pricing is less favorable than the market assumes. A sharp move lower in NAND spot prices would likely hit sentiment first, before the contract backlog can fully offset it.

Contrarian view: the market may be too focused on “cheap on FY27 earnings” and not enough on terminal margin durability. If a meaningful share of production is now contracted, the real upside is not another year of parabolic EPS, but a lower earnings volatility profile that could justify a permanently higher base multiple. That makes the stock attractive for investors who want to own the normalization of a cyclical into a quasi-annuity, but less compelling for those underwriting continued commodity-style torque.

Key Decisions for Investors

• Maintain/establish a tactical long in SNDK for 1-3 months into the next contract-signing cadence; target continued multiple expansion if management shows additional NBM wins. Use a trailing stop if NAND spot prices roll over or if the market starts pricing in rapid supply response.
• Pair trade: long SNDK vs short a weaker memory/module name or broad semiconductor ETF hedge over 3-6 months. The thesis is that contract-backed visibility should de-risk SNDK relative to more exposed peers while the sector remains sensitive to any normalization scare.
• Buy SNDK call spreads 3-6 months out rather than outright calls to express upside to rerating while capping premium risk; this is best if you expect a slower grind higher rather than another explosive leg.
• Take profits selectively on any sharp post-earnings pop above current implied fair value; the risk/reward deteriorates if the market fully discounts the contract backlog before margin durability is proven.
• Monitor hyperscaler AI capex and NAND spot pricing monthly; if either weakens, rotate from long SNDK to a hedged posture quickly, as the stock’s beta to sentiment remains high despite the new contracts.