Sandisk stock forecast: Kioxia’s 11% Tokyo slide tests NAND growth

Tokyo’s first trading hour on Monday delivered a sharp warning to investors awaiting the next U.S. session in Sandisk. Kioxia Holdings, the Japanese flash-memory maker that supplies substantially all of Sandisk’s flash wafers through their joint ventures, traded at ¥11,985 at 9:34 a.m. Japan time, down 11.0% from Friday’s ¥13,470 close, according to a delayed Yahoo Finance quote. Sandisk shares had already fallen 13.4% in Friday’s U.S. session on Alpaca’s IEX closing series. Those are observations in two different markets and sessions, not a prediction that Sandisk will repeat Kioxia’s Monday move.

The useful forecast question is narrower: does Kioxia’s share-price slide signal that the pricing surge in flash memory has reached a plateau, or is Sandisk’s more ambitious near-term revenue guide still credible? TECHi’s calculation from the companies’ published figures exposes the gap. Kioxia expects revenue to rise just 0.8% sequentially in its October–December quarter, from ¥2.480 trillion to ¥2.500 trillion. Sandisk guides to a 12.4% sequential increase at the midpoint of its next-quarter range, from $11.031 billion to $12.400 billion. The difference is 11.6 percentage points. It is a comparison of growth rates, not a conversion of yen into dollars or an assertion that the two companies sell the same product mix.

Article Brief

Key Takeaways

4 Points24s Read

• 01Tokyo signal–Kioxia traded at ¥11,985 at 9:34 a.m. JST November 2, 11.0% below its prior close, on a delayed quote.
• 02The gap–Kioxia guides to 0.8% sequential revenue growth, versus 12.4% at Sandisk’s guidance midpoint, an 11.6 percentage-point difference.
• 03Shared wafers–Sandisk says substantially all of its flash wafers come from manufacturing ventures shared with Kioxia.
• 04The forecast test–Sandisk’s November U.S. session and later price, mix and margin disclosures will show whether the Tokyo decline reads across.

Why Kioxia’s fall matters to Sandisk

The two companies are closer than ordinary semiconductor peers. Sandisk says in its latest Form 10-Q that it procures substantially all its flash-based memory wafers from three manufacturing ventures with Kioxia. The filing also describes facility arrangements, prepayments and research commitments tied to those ventures. They share a production base, although they do not have identical customers, sales channels or financial statements. A shift in wafer economics at one partner therefore deserves attention at the other; it does not mechanically transfer one company’s stock return to the other.

The new signal is the market’s first reaction after Kioxia’s October 30 results. The Kioxia financial release reported September-quarter revenue of ¥2.480 trillion, up from ¥1.767 trillion in the June quarter. That is a 40.3% sequential increase, driven, the company said, by higher average selling prices, AI data-center demand and more bits shipped. Yet its next-quarter outlook is ¥2.500 trillion of revenue and ¥1.995 trillion of operating profit, essentially level with the latest quarter’s ¥2.480 trillion and ¥1.995 trillion. Kioxia says strong data-center demand should continue, while exchange rates could weigh on the next three months. Flat guidance after exceptional growth is not the same as collapsing demand. It does, however, change the incremental growth question investors must answer.

Sandisk’s October 29 earnings release shows a different trajectory. Revenue grew 23% sequentially to $11.031 billion in its fiscal first quarter, and the company projected $12.2 billion to $12.6 billion for fiscal second quarter. The midpoint is $12.4 billion; dividing the $1.369 billion increase by the $11.031 billion starting figure gives 12.4%. The low and high ends imply growth of 10.6% and 14.2%, respectively. Sandisk’s data-center revenue rose 61% sequentially to $4.802 billion, while consumer revenue declined 29% to $397 million. This mix is one reason a flat forecast at Kioxia need not invalidate Sandisk’s guidance.

The shared supply chain and the unshared revenue line

The joint-venture link is the standout feature of this comparison. Sandisk and Kioxia do not merely observe the same NAND market from afar; they use common wafer ventures. But the Kioxia filing splits revenue among SSD and Storage, Smart Devices and Other, while Sandisk reports Datacenter, Edge and Consumer end markets. These categories overlap without matching. Kioxia’s reported SSD and Storage line includes PC, data-center and enterprise products. Its Other line includes sales to the Sandisk group through the ventures. Sandisk can sell a different mix of finished products, on different commercial terms, into the same quarter.

The accounting calendars also differ slightly. Kioxia’s forecast covers October 1 through December 31, 2026. Sandisk’s guidance is for its fiscal second quarter following a quarter that ended October 2. The two outlooks largely cover the same demand window but are not perfectly aligned. Nor should operating profit margins be compared directly without reconciling IFRS, U.S. GAAP, non-GAAP adjustments and each company’s venture accounting. The more defensible cross-check is sequential revenue growth within each company, then an explanation of why that growth may differ.

Kioxia’s reported June-to-September revenue jump was about ¥712.9 billion, or 40.3% on the ¥1.7671 trillion starting quarter. Its October-to-December forecast adds only ¥20 billion, or 0.8%. Sandisk’s prior revenue increase was $2.066 billion, or 23.0% on $8.965 billion; its guidance midpoint adds another $1.369 billion, or 12.4%. Both imply deceleration, but Kioxia’s is far steeper. That difference is an early warning to test Sandisk’s assumptions on selling prices, data-center mix and customer commitments when its next numbers arrive.

What the Tokyo price is actually saying

At the measured 9:34 a.m. Japan time quote, Kioxia’s ¥11,985 price was ¥1,485 below Friday’s ¥13,470 close. The calculation is ¥1,485 divided by ¥13,470, or 11.0%. The delayed quote is a fixed intraday snapshot; the Japanese session can move after that moment. At about the same point, Tokyo Electron, a chip-equipment supplier with a different business model, traded at ¥10,135, up 6.5% from ¥9,514. Its October 30 forecast called for fiscal-year sales growth of 51.4% and operating-income growth of 87.1%, explicitly citing AI-server investment. The 17.5-percentage-point gap between those two Japanese share moves shows how selectively investors were treating AI-linked semiconductor exposures at the open. It does not isolate the cause of either move.

Sandisk’s last observed U.S. close predates this Tokyo action. Alpaca’s IEX daily bars show $1,330.74 on October 29 and $1,152.04 on October 30, a 13.4% close-to-close fall on that venue. The wider U.S. market was also under pressure Friday, so attributing Sandisk’s entire decline to its own outlook would go beyond the evidence. The next direct test is the U.S. cash market open on November 2, when traders can respond to the fresh Kioxia signal. No U.S. reaction is assumed before it happens. Readers can follow SNDK’s TECHi quote page for the later session rather than treating Friday’s close as a live Monday price.

Sandisk stock forecast for November: three testable paths

Base case: Sandisk’s $12.2 billion to $12.6 billion fiscal-second-quarter revenue guide remains plausible despite Kioxia’s flat outlook. The common wafer base can coexist with different end-market exposure, contract pricing and shipment timing. Under this case, the Tokyo drop raises scrutiny but does not by itself change Sandisk’s reported guidance. The stock’s near-term path would depend on whether U.S. investors see Kioxia as a market-wide pricing warning or a company-specific forecast shaped by currency and mix.

Upside case: The company provides evidence that data-center demand and contracted pricing protect its revenue growth even as Kioxia’s company-wide sales flatten. Sandisk’s 61% sequential data-center revenue growth in the reported quarter is an encouraging historical figure, but it is not proof that the same rate persists. A sustained recovery after the U.S. open, paired with later disclosure that next-quarter data-center sales and gross margin hold up, would strengthen this case. The SNDK forecast page is a useful place to compare any subsequent analyst revisions, with their dates and methods, against the company’s own guidance.

Downside case: Kioxia’s near-flat outlook is the first visible sign that the unusually rapid rise in NAND average selling prices is slowing across the shared production ecosystem. If Sandisk’s next reported revenue approaches or misses the $12.2 billion low end, or its expected 84.5%–85.5% non-GAAP gross-margin band slips, the market would have firmer evidence that recent profitability was closer to a peak than a durable step up. The condition is observable in future results; Kioxia’s share price alone does not establish it.

These are conditional scenarios, not price targets. Sandisk also announced an eight-for-one split effective later in November, according to its earnings release. A split changes the per-share quotation, not the underlying market value or the business test. The company’s stated non-GAAP outlook also excludes certain stock-based compensation, so readers should compare GAAP and adjusted profitability when the next filing is available.

The evidence that would change the call

The strongest counterargument to a bearish read-through is in Kioxia’s own document: it still expects demand from data centers to be strong and identifies exchange rates as a reason its next-quarter revenue and profit may be broadly flat. Sandisk’s 10-Q, meanwhile, says AI-driven demand is expected to persist through calendar 2027 and describes longer-term customer agreements. Those are management expectations and contractual descriptions, not a guarantee of future sales or pricing. A sharply falling stock can be reacting to expectations that were more demanding than the published forecast, rather than to evidence of a collapsing NAND market.

For November, the most useful checks are precise. First, observe Sandisk’s actual U.S. opening and closing prices on November 2, including provider, time and session; do not substitute Kioxia’s percentage move. Second, watch whether Sandisk or Kioxia clarifies how much of its next-quarter growth is price, bits shipped, customer mix or foreign exchange. Third, examine whether further filings show expansion or strain in the shared manufacturing commitments. Sandisk’s 10-Q says its three Flash Ventures run through 2034 and notes an $838 million equivalent share of remaining lease guarantees as of October 2. Those are real balance-sheet and capacity considerations, not a one-day trading signal.

The forecast therefore remains conditional: Sandisk’s published growth guide has room to be met, but Kioxia’s opening plunge makes the gap between their near-term revenue expectations the question to verify. If Sandisk sustains double-digit sequential revenue growth with strong data-center mix, the Tokyo selloff may prove a poor proxy for SNDK. If pricing or margins soften across both partners, the shared-wafer connection will look more predictive. The next U.S. session is the first chance to observe that distinction, not the final verdict.