A stock that has risen 35-fold in a year cannot be analysed like a semiconductor company. SanDisk (SNDK) closed at $1,641.11 on August 14, 2026, up 7.39% on a session in which the entire memory complex moved together, having traded at $46.68 twelve months earlier — a gain of 3,415.7% (stockanalysis.com daily close, retrieved August 15, 2026). It is also 30.3% below its 52-week high of $2,354.39, set on June 22, 2026. Both of those facts describe the same twelve months, and holding them together is the only honest way to value this business.
Here is the detail that decides the case, and it is buried in the company’s own disclosure rather than in any headline. Of SanDisk’s sequential revenue growth in the most recent quarter, roughly two-thirds came from pricing and only one-third from volume. That single ratio reframes everything. A business growing on volume is taking share; a business growing on price is riding a shortage. Shortages end — not because the company does anything wrong, but because price is the mechanism by which supply is eventually summoned. The bear case for SanDisk does not require a single operational misstep. It requires only that NAND pricing normalises, and the company has already told us that pricing is doing most of the work.
Key facts
- Spot price $1,641.11, close of August 14, 2026, +7.39% on the day — stockanalysis.com, retrieved August 15, 2026
- One year earlier (August 14, 2025) the stock closed at $46.68 — a gain of 3,415.7%
- 52-week high $2,354.39 (June 22, 2026); 52-week low $42.82 (August 15, 2025)
- Q4 FY2026 revenue $8.97 billion, up 372% year on year; adjusted EPS $39.25 against a $34.52 estimate
- Q4 FY2026 gross margin 84.6% — a company record
- Full-year FY2026 revenue $20.25 billion; net income $11.43 billion
- Q1 FY2027 guidance $10.3–10.8 billion at 83–85% gross margin
- Long-term contracts: approximately $93.9 billion of minimum commitments across 8 contracts with 6 customers on four-year terms
The sector move matters, and it cuts both ways
The 7.39% gain on August 14 was not a SanDisk event. The whole memory and storage complex rose together, and a sector-wide move is materially weaker evidence about any single company than a stock-specific one. When correlated names move in lockstep, the market is repricing a commodity — NAND — rather than reassessing a management team.
That distinction is worth holding onto, because it works in both directions. It means SanDisk’s gain that day tells you very little about SanDisk’s execution. It also means the 30.3% drawdown from the June high is not a verdict on SanDisk either. Both moves are the cycle talking. FinanceFeeds has tracked the same complex through Micron and Western Digital, the company from which SanDisk was separated, and the correlation across those names is the clearest available evidence that this is a commodity repricing rather than a set of independent company stories.
What the Investor Day actually established
SanDisk’s Investor Day produced the most specific forward guidance the company has given. Chief Financial Officer Luis Visoso put the NAND market above $300 billion in 2026 and above $500 billion in 2027, with supply tightness extending into 2028. The FY2028–FY2030 model targets mid-to-high-teens annual revenue growth, non-GAAP gross margins around 80%, and adjusted free-cash-flow margins near 50%.
The strongest single datapoint for the bull case is the contract book: approximately $93.9 billion of minimum commitments across eight contracts with six customers, on four-year terms. Against FY2026 revenue of $20.25 billion, that is more than four years of current revenue contracted as a floor. Minimum commitments are not the same as recognised revenue, and the pricing within them is not disclosed — but as downside protection goes, it is unusually concrete.
The data-centre mix shift supports the same argument. Q4 data-centre revenue reached $2.98 billion, doubling sequentially and up 645% year on year, with SanDisk’s bit shipment share rising from roughly 12% to about 38% within a year. That is a genuine change in the customer base, not a price effect, and it is the part of the story that would survive a pricing normalisation. FinanceFeeds covered the event in its analysis of SanDisk’s answer to the peak-cycle fear.
The supply-side evidence is genuinely tight
The most persuasive support for SanDisk’s supply-tightness argument came from outside the company. Applied Materials, reporting on August 13, posted record revenue of $9.12 billion, up 25% year on year, with Semiconductor Systems at $7 billion. Its NAND commentary was the relevant part: wafer starts declining, and capital expenditure focused on upgrades rather than new capacity.
That is the equipment supplier — the company that would be first to see a supply response — reporting that the supply response is not being built. In a commodity cycle, that is the single most important variable, because shortages persist exactly as long as capacity additions lag demand. It is a meaningful, independent corroboration of the CFO’s 2028 claim.
Elon Musk’s observation in August 2026 that “memory is AI’s biggest bottleneck” is directionally consistent, though it is commentary rather than data and should be weighted accordingly.
The peak-cycle problem the contracts do not solve
Now the other side, stated as strongly as it deserves. An 84.6% gross margin is not a normal state for a memory business. Historically, NAND has been among the most brutally cyclical products in the semiconductor industry, with gross margins that swing from the seventies to negative territory across a cycle. SanDisk’s own FY2028–FY2030 model assumes approximately 80% — but the framing question is whether an 80% steady-state margin in a commodity industry is a plan or an aspiration.
The pricing-versus-volume split is the tell. If two-thirds of sequential growth is price, then a normalisation in NAND pricing does not merely slow growth — it reverses a large part of it, and it does so at the gross-margin line first, where the effect on earnings is amplified. The $93.9 billion contract book provides volume protection. It does not obviously provide price protection, because the contracted pricing is not disclosed.
Supply risk from China compounds this. FinanceFeeds examined CXMT’s Shanghai debut and its implications for the memory complex in its analysis of the listing. New Chinese capacity is precisely the supply response that ends shortages, and it arrives on a political timetable rather than an economic one.
What the FY2030 model is really claiming
SanDisk’s own long-range model deserves closer reading than it usually gets, because it contains an implicit concession. For FY2028–FY2030 the company guides to mid-to-high-teens annual revenue growth, non-GAAP gross margins around 80%, and adjusted free-cash-flow margins near 50%.
Set that against what the company is doing now. FY2026 delivered revenue growth of a wholly different order and a Q4 gross margin of 84.6%. The model is therefore not an extrapolation of the present — it is a guide to substantial deceleration, from a hyper-growth year to mid-to-high teens, with margins stepping down from 84.6% to roughly 80%. Management is telling investors the current run-rate is not the steady state.
That is creditable and it is also the crux. A 50% adjusted free-cash-flow margin sustained through FY2030 would be remarkable for any hardware business and unprecedented for a NAND producer across a full cycle. If it holds, today’s price is defensible on cash generation alone. If NAND behaves as NAND has always behaved, the margin assumption is the first thing to break, and the model’s mid-to-high-teens growth compounds off a much lower base than the guide implies.
The separation from Western Digital sharpens the point. SanDisk is now a pure-play NAND business, without the hard-disk-drive segment that historically dampened the swing in either direction. That focus is precisely why the stock rose 35-fold, and it is precisely why the downside is less cushioned than it would have been inside the combined company.
Where the analysts sit
Consensus across 23 analysts compiled by S&P Global stood at $2,094, with a range from $1,000 at the bearish end to $3,600 at the bullish, on a Buy consensus. J.P. Morgan moved to Overweight with a $2,250 target following the Investor Day. Bernstein carried $3,000 from late June, citing the contract book’s downside protection.
Two things stand out. The spread — $1,000 to $3,600, a 3.6x range — is extraordinarily wide for a covered large-cap, and it says the analyst community is no more settled on the terminal value than the share price has been. And the consensus of $2,094 sits 27.6% above the current $1,641.11, meaning the street collectively regards the 30.3% drawdown from June as an overreaction.
Bull case: $2,400
The bull case is that supply tightness genuinely extends into 2028, as both the company and its equipment supplier now indicate, and that the data-centre mix shift from 12% to 38% of bit shipments proves durable. On that path, Q1 FY2027 guidance of $10.3–10.8 billion is a waypoint rather than a peak, and the $93.9 billion contract book converts steadily.
A target of $2,400 is approximately 46% above spot and marginally above the 52-week high of $2,354.39. It sits modestly above the $2,094 consensus and well below Bernstein’s $3,000, which is the appropriate place for a case that requires the cycle to extend but not to re-accelerate to new extremes. Critically, it requires no expansion in the multiple — only that the guided revenue and margin profile is delivered while the cycle holds.
Bear case: $1,150
The bear case needs nothing from management. It needs NAND pricing to normalise. If the two-thirds of sequential growth attributable to price reverses even partially, revenue growth decelerates sharply and the 84.6% gross margin compresses toward historical norms — and because those two effects compound, earnings fall considerably faster than revenue.
The catalysts are identifiable: Chinese capacity additions arriving ahead of schedule, hyperscale customers digesting after a period of aggressive ordering, or simply the arithmetic of a market the CFO expects to grow from $300 billion to $500 billion attracting the capital that ends every shortage.
A target of $1,150 is roughly 30% below spot and sits above the $1,000 analyst bear low. It does not contemplate a return toward the $42.82 low of a year ago, because the business today bears no resemblance to the business then — FY2026 revenue of $20.25 billion and net income of $11.43 billion are real. It contemplates a cycle turning, which is the base rate for this industry.
The technology roadmap is the quiet advantage
One element of the Investor Day gets less attention than the contract book but may matter more over a full cycle. SanDisk’s ninth-generation 2Tb QLC 3D flash carries a 4.8 Gb/s interface, a 33% improvement, and tenth-generation NAND delivering 59% higher bit density than its predecessor entered production in July at the Kitakami Fab2 facility. The company also disclosed early technical specifications for High Bandwidth Flash through the Open Compute Project, with Google participating.
Bit density is the variable that determines cost per gigabyte, and cost per gigabyte is what decides who survives a downcycle. A producer entering a price normalisation with a 59% density advantage over its own prior generation is structurally better placed than one that is not, because the marginal cost floor falls with density. This does not prevent the cycle from turning. It changes who remains profitable when it does — which is the distinction between the bear case being a drawdown and the bear case being an impairment.
What to watch
Three signals rank above all others. First, the pricing-versus-volume split in the next quarterly disclosure: if the volume share rises above one-third, the growth is becoming more durable, and if it falls further, the opposite. Second, gross margin against the 83–85% guide, since that is where pricing normalisation appears first. Third, any change in Applied Materials’ NAND commentary — a shift from declining wafer starts and upgrade-focused capex toward new capacity would be the earliest available warning that the supply response has begun.
SanDisk has executed exceptionally and the contract book is a genuine asset. But at 35 times its price of a year ago, the share price is a wager on where the NAND cycle turns, not on how well the company is run. Those are different questions, and only one of them is within management’s control.
Frequently asked questions
How much has SanDisk stock risen in the past year?
SanDisk closed at $1,641.11 on August 14, 2026, against $46.68 on August 14, 2025 — a gain of 3,415.7%, or roughly 35 times. Despite that, it trades 30.3% below its 52-week high of $2,354.39 set on June 22, 2026.
Why did SanDisk stock rise 7% on August 14?
The gain came as the entire memory and storage complex moved higher together rather than on a SanDisk-specific catalyst. Sector-wide moves reflect a repricing of NAND as a commodity and say less about any individual company’s execution than stock-specific moves do.
Is SanDisk’s growth driven by price or volume?
Predominantly price. The company disclosed that roughly two-thirds of its sequential revenue growth came from pricing and one-third from volume. That matters because pricing-led growth reverses when a shortage ends, whereas volume-led growth reflects durable share gains.
What are analysts’ price targets for SNDK?
Consensus across 23 analysts compiled by S&P Global was $2,094, ranging from $1,000 to $3,600 on a Buy consensus. J.P. Morgan moved to Overweight at $2,250 after the Investor Day, and Bernstein carried $3,000 from late June.
What are the bull and bear targets for SanDisk?
This analysis sets a bull case of $2,400, about 46% above the $1,641.11 spot, requiring supply tightness to extend into 2028 as guided. The bear case is $1,150, roughly 30% below spot, requiring only that NAND pricing normalises and gross margins compress from 84.6% toward historical levels.
This article is informational analysis only and is not financial, investment, or trading advice. Equity markets are volatile and prices can fall as well as rise. Past performance does not guarantee future results. Do your own research and consult a regulated financial adviser before making any investment decision.
