SanDisk is the hardest of the big memory names to value, and the reason is structural rather than mathematical. It is a pure NAND flash maker, it was spun out of Western Digital only in 2025, and it has been a standalone public company for barely more than a year. So you are trying to estimate the through-cycle earnings of a deeply cyclical business that has almost no standalone cycle on record. That uncertainty should sit at the center of any SanDisk valuation.
This article applies the cyclical framework from how to value memory and cyclical stocks to SanDisk specifically, and it pairs naturally with the Micron valuation, which faces the same cycle from a more diversified position.
What SanDisk Is After the Spinoff
SanDisk became an independent company when Western Digital separated its flash business, with the separation completed on February 21, 2025. What remains is close to a pure play on NAND flash, the non-volatile memory used in solid-state drives, memory cards, and the storage tier of data centers.
That purity is the key fact for valuation, and it cuts in one direction. Micron makes both DRAM and NAND and carries the high-bandwidth memory used in AI accelerators, which spreads its exposure across products at different points in their own cycles. SanDisk has no DRAM and no high-bandwidth memory. It is concentrated in the single most commoditized corner of the memory market. When NAND pricing is strong it earns very high margins, and when NAND pricing breaks it has nothing else to lean on. There is no diversification cushion here.
SanDisk does sell into a range of end markets, and it breaks them out, which helps. But every one of them buys the same underlying commodity, so the end-market split tells you where demand is coming from, not whether the business is insulated from the NAND price.
The Latest Quarter
SanDisk's most recent reported quarter, FQ3 of fiscal 2026 ended April 3, 2026, shows the upswing in full force. These figures are from the company's SEC earnings filing.
| Metric | FQ3 FY2026 | Year earlier |
|---|---|---|
| Revenue | $5.95B | $1.70B |
| Revenue growth | +251% year over year | |
| Gross margin | 78.4% | |
| Non-GAAP diluted EPS | $23.41 | |
| Operating cash flow | $3.04B | |
| Free cash flow | $2.99B | |
| Long-term debt | $0 |
A gross margin near 78% on a NAND commodity is not a normal state of the world. It is what the top of an extreme shortage looks like. The end-market detail shows how concentrated the surge is in AI-adjacent demand:
| End market | Revenue | Year-over-year growth |
|---|---|---|
| Datacenter | $1.47B | +645% |
| Edge | $3.66B | +295% |
| Consumer | $0.82B | +44% |
Datacenter revenue grew more than sevenfold in a year. SanDisk also reported signing five longer-term supply agreements under a new commercial model, which management presents as evidence that some of this demand is being locked in rather than spot. Even so, the company guided the next quarter higher again, to roughly $7.75B to $8.25B in revenue, which only sharpens the question every cyclical eventually poses: how much of this holds.
Why the Valuation Is Especially Hard Here
As of late May 2026, SanDisk traded near $1,500 a share, up roughly fivefold year to date, for a market capitalization around $225 billion. The forward price-to-earnings ratio sat in the high single digits, the same deceptively low multiple that Micron carries and for the same reason: the denominator is a peak earnings forecast. The logic is in , and it applies with extra force to a pure-play at the top of its cycle.
SanDisk adds a problem Micron does not have. To normalize earnings you average across a full cycle, and SanDisk has been standalone for only about fifteen months. There is no multi-cycle record of its margins, its capital spending, or its trough behavior as an independent company. You can lean on the broader NAND industry history and on its years inside Western Digital, but the error bar around any normalized figure is wider than for an established standalone like Micron. When the history is thin, the discipline is to widen the range and demand more margin of safety, not to narrow it.
A Normalized Valuation Framework
The same method applies, with the caveat that every input is more uncertain. The table below is illustrative. It shows how SanDisk's intrinsic value moves with the normalized free cash flow per share you assume, using a standard two-stage discounted cash flow.
| Scenario | Normalized FCF per share | Growth, years 1-5 | Discount rate | Illustrative DCF value |
|---|---|---|---|---|
| Bear | $8 | 2% | 12% | roughly $95 |
| Base | $20 | 4% | 11% | roughly $275 |
| Bull | $40 | 6% | 10% | roughly $650 |
For reference, annualizing the most recent quarter's free cash flow implies something like $75 per share at the current peak run rate. The scenarios above all sit below that, because they are estimates of the mid-cycle figure rather than the peak. And all three sit well below the late-May share price near $1,500.
As with Micron, that gap is not an automatic verdict that the stock is overvalued. It is a statement of what the price assumes. To justify the current price on a discounted cash flow you would need to believe the normalized free cash flow is well above what past NAND cycles produced, whether because AI keeps flash demand elevated for years, because supply stays tight longer than usual, or because the right discount rate is lower than NAND's history would suggest. That is a real possibility and a real debate. The framework's job is to force the assumption into the open so you can decide whether you believe it, rather than backing into it from a low headline multiple.
The Risks Are Real
SanDisk carries the highest cycle risk of the memory names covered here, and it is worth being blunt about why. It is a single-product commodity maker with no diversification, valued on peak margins, with a short standalone history that makes its true mid-cycle earnings genuinely hard to estimate. NAND has historically been the more brutal half of the memory market, with deeper and more frequent margin collapses than DRAM, including periods of negative gross margin across the industry.
If NAND pricing turns, SanDisk has no second business to absorb the blow, and free cash flow can swing from strongly positive to negative within a few quarters. The zero-debt balance sheet is a genuine strength that gives it room to survive a downturn, but survival and value are different questions. This is the kind of situation where is not a nicety but the core of the analysis. The wider the range of outcomes, the larger the discount to your central estimate you should require before the risk is worth taking.
How to Run Your Own SanDisk Valuation
There is no clean way to hand you a SanDisk price target, and you should be skeptical of anyone who offers one with confidence given how little standalone history exists. The useful output is a range, and the range depends almost entirely on the normalized free cash flow you are willing to defend.
The best way to engage with it is to build the scenario yourself. Open the , enter a mid-cycle free cash flow rather than the current peak run rate, and test how sensitive the intrinsic value is to that one assumption. Then widen the range to reflect the thin standalone history, and see whether the current price still fits inside any version you find defensible.
Frequently Asked Questions
Why is SanDisk harder to value than Micron? SanDisk is a pure NAND maker with no DRAM or high-bandwidth memory to diversify it, and it has been a standalone public company only since 2025. With no multi-cycle standalone record, its normalized through-cycle earnings are harder to estimate, so the range of plausible values is wider than for a diversified, established memory maker.
What does SanDisk make? SanDisk makes NAND flash memory, the non-volatile storage used in solid-state drives, memory cards, and the storage tier of data centers. It does not make DRAM or high-bandwidth memory, so it is concentrated in the most commoditized part of the memory market.
Is SanDisk's low forward PE a sign it is cheap? Not on its own. The forward PE divides the price by next year's expected earnings, and those earnings are forecast near a cycle peak. A peak denominator makes the multiple look low even when the price already reflects an expected decline. For a pure-play cyclical this is a warning to investigate, not a bargain to act on.
Does the zero-debt balance sheet make SanDisk safe? It helps. A debt-free balance sheet gives SanDisk room to survive a NAND downturn without financial distress. But survival is different from value. The stock can still fall a long way if peak earnings normalize, regardless of how strong the balance sheet is.
Ready to build your own scenario? Open the and enter a normalized free cash flow instead of the peak figure.
