Sandisk Corporation SNDK
Sandisk's FY2026 results (fiscal year ended 2026-07-03) show an extraordinary turnaround—revenue of $20.25B with a 71.5% gross margin, $11.43B net income, and $11.49B free cash flow, versus a $1.64B net loss in FY2025. The P/E of 22.97 screens cheap against a peer median of 51.53, and the AI-storage thesis is reinforced by datacenter revenue growth and $93.9B in contracted revenue commitments cited at Investor Day.
However, the stock has already rallied over 3,400% in 12 months, sits 30% below its peak, insiders have net-sold $817M over 24 months with zero cluster buying, and NAND's cyclical margins mean today's 61.2% operating margin is unlikely to be sustained through a full cycle—making this a hold pending evidence that current pricing power persists into FY2027.
What could go wrong
- NAND cyclicality. FY2026 operating margin of 61.2% and gross margin of 71.5% are cyclical peaks; FY2023 saw operating margin of -33.4% when NAND pricing collapsed, and a similar downcycle would compress earnings sharply.
- Insider selling. Over 24 months insiders net-sold 19.7M shares worth $817M with 66 dispositions versus 18 acquisitions and no cluster buying, suggesting limited insider conviction at current levels.
- Valuation on revenue. P/S of 13.07 is roughly 4x the peer median of 3.44, so the stock is expensive on sales even if the P/E of 22.97 looks cheap—earnings would need to hold near peak to justify the price.
- Momentum cooling. The stock is down 30% from its peak and memory stocks fell ~5% on 2026-08-18 premarket, signaling that the crowded AI-memory trade may be unwinding.
What would change my mind
Where this comes from: FMP FY2026 annual (period ended 2026-07-03) · FMP FY2025 annual (period ended 2025-06-27) · peer_relative composite · insider summary (24-month window). Orin's read on SNDK; not advice.
Twelve months actual closes to 2026-09-23 · actual filings
50-day average 200-day average · volume below
On file
All SNDK filingsLargest holders · 13F
Money flowNo 13F positioning on file.
Insiders, last 24 months: 174 Form 4 filings, net −$1.1B. Of the 50 on hand, 0 were open-market purchases and 39 sales— the rest are grants, option exercises and tax withholding.
Ask Orin about SNDK
Orin answers questions about SNDK from its filings, transcripts and numbers — with the source behind every claim. Your own desk starts with 3 free days, card required.
Start 3 free days →Where it trades
| Metric | Now | Own median | Sector |
|---|---|---|---|
| P/E | 23.4× | — | 53.3× |
| EV/EBITDA | 20.2× | — | — |
| P/S | 13.29× | 0.81× | — |
| P/B | 17.0× | 0.5× | — |
Its P/E sits above all 4 of the last 4 years (+5.60σ from its own mean).
9.8%
free-cash-flow growth every year for 10 years, discounted at 10%, is what today’s price implies — starting from $11.5B of trailing free cash flow. It is an assumption the market is making, not a forecast Orin is making.
What Wall Street published
$2153
$1200 – $3050 · +19% against today's price
- 14 buy or overweight
- 2 hold
- 0 underweight or sell
The analysts’ consensus and their own ratings, not Orin’s. A target is a twelve-month opinion published by the firm that wrote it.
Against the companies it is compared to
| Company | Market cap | P/E | EV/EBITDA | Gross | Net | ROE |
|---|---|---|---|---|---|---|
| SNDKSandisk Corporation | $269B | 23.4× | 20.2× | 71.5% | 56.5% | 93% |
| CRDOCredo Technology Group Holding Ltd | $36B | 65.7× | 62.7× | 67.1% | 33.8% | 27% |
| FLEXFlex Ltd. | $41B | 43.4× | 23.4× | 9.5% | 3.3% | 19% |
| HPEHewlett Packard Enterprise Company | $83B | 31.0× | 17.1× | 36.0% | 6.6% | 11% |
| JBLJabil Inc. | $32B | 38.1× | 16.4× | 9.2% | 2.6% | 62% |
| KEYSKeysight Technologies, Inc. | $60B | 48.7× | 35.3× | 64.7% | 19.1% | 20% |
| SMCISuper Micro Computer, Inc. | $27B | 11.4× | 8.5× | 10.8% | 5.7% | 25% |
| TDYTeledyne Technologies Incorporated | $29B | 29.9× | 19.2× | 39.0% | 15.3% | 9% |
| TERTeradyne, Inc. | $61B | 53.1× | 41.7× | 59.3% | 25.8% | 38% |
The median is of the 8 peers listed above and nothing else — check it against the column. This company trades 42.7% below it. The peer set is the vendor’s, so a company can appear here that a person would not have chosen.
What it has traded at
What its sector has traded at
Two scales, drawn separately on purpose: a sector median and one company’s multiple are not the same number and putting them on one axis invites a comparison neither series supports.
Discounted cash flow
| Model | Fair value | Against | Computed |
|---|---|---|---|
| Standard DCF | $49 | $1643.00 · −97% | 2026-06-10 |
| Levered DCF | $24 | $1643.00 · −99% | 2026-06-10 |
The vendor’s intrinsic-value models. Treat them as a low-growth anchor — for a fast grower the market routinely pays well above a DCF. These were computed against the price in the “against” column, not today’s.
Every estimate on one scale
The vertical rule is the current price. Ranges and point estimates are the backend’s, published as they were calculated; an arrow means the estimate sits beyond this scale, and its number is on the right.