SYNTHOS RESEARCH

Sandisk SNDK

Technology · Hardware, Equipment & Parts · Synthos Deep Dive · 2026-07-03

$1,745.00
Hold
Risk 9Growth 5Exponential 3Fair value $984 $450–$1443

At a glance

VerdictHold — systematic Synthos tier
Price (2026-07-03)$1,745 (−14.1% on the day) · market cap ~$258B
Synthos scores (0–10)Downside Risk 9 · Growth Quality 5 · Exponential Potential 3
Synthos fair value (base case)~$984−44% · full range $450 (bear) – $1,443 (bull)
Street consensus$1,684 (high $3,000 / low $450; 13 Buy · 2 Hold) — context, not our anchor; note it sits above our bull
Valuation57× trailing EPS · 27× FY26E · EV/S 19.3× · EV/EBITDA 47× — priced on peak-cycle earnings
Exponential Potential3/10 · Low — the surge is a NAND up-cycle spike, not secular acceleration; the estimate curve itself shows a FY30 earnings collapse
TechnicalsParabolic — $1,745, up ~38× in 12 mo (+3,781%), −25% off the 52-wk high, RSI 47, beta 4.88
ConvictionLow — 1 KB voice, bearish (compound_and_friends, conviction 72); no net-bullish expert coverage
Position sizingAvoid / 0% for core holders; speculative-only, and then tiny, for cycle traders
Next catalyst2026-08-13 Q4 FY26 earnings (Street EPS $33.64)
Single biggest riskThe NAND cycle rolls over — memory earnings swing to losses, and the multiple de-rates on collapsing EPS at once

One-line thesis. Sandisk is a good company caught in a violent NAND memory up-cycle that has re-rated the stock ~38× in twelve months; the earnings are real today (Q3 FY26 EPS $24.43, 78% gross margin) but they are peak-cycle, the balance sheet was posting a $1.6B loss just a year ago, beta is 4.88, and at $1,745 the market is capitalizing the top of the cycle — we see through-cycle fair value near $984 and rate it Avoid.

◆ Synthos call — Hold SNDK is a solid business largely reflected at ~$984 — fine to keep, no reason to chase; it gets interesting again below ~$836.
Downside Risk (lower = safer)
9/10 · Very High
Beta 4.88, memory cyclical up ~38× in 12 mo, priced on peak-cycle EPS — the single riskiest name in the pool.
Growth Quality
5/10 · Moderate
Explosive TTM upturn (56% GM, 34% net) is real but cyclical, not durable; FY25 was a $1.6B loss.
Exponential Potential
3/10 · Low
Growth is a memory up-cycle spike, not secular acceleration; estimates themselves show a FY30 collapse.
⚖ Reverse-DCF cross-check Market-implied growth ≈ 39%/yr To justify today’s $1,745, earnings would have to compound roughly 39% a year for 10 years (9% discount rate). Analysts forecast ~80%/yr, so the market is pricing in LESS than what the Street expects.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
Buy — CoreOwn it as a foundation — start or add now, size it for years, let dips be gifts.
Buy — TacticalGood price + confirmed trend + a defined exit — buy the setup, not a marriage.
WatchWe want the business, just not at this price/setup — act only when the listed trigger hits.
HoldFine to keep if you own it — no reason to buy more; new money does better elsewhere.
AvoidDon't own it — the problem is the business or the expectations, so a cheaper price won't fix it.

In plain English

Sandisk makes flash-memory storage — the chips inside SSDs, memory cards, USB drives, and data-center storage. Memory is one of the most boom-and-bust businesses in the world: prices and profits swing wildly. Right now we are in a boom — an AI-driven memory shortage sent prices soaring, and Sandisk swung from a big loss a year ago to huge profits today. The stock followed: it is up roughly 38 times over the past year.

The catch: you are buying at the top of the boom, at a top-of-the-boom price. When the memory cycle turns — and it always does — profits can vanish and the stock can fall hard. Our verdict is Avoid: this is a great trade that already happened, not a durable long-term holding to buy today.

Here is what our three scores mean in everyday terms:

The one big worry: the memory cycle rolls over. When NAND prices fall, Sandisk's profits can swing back to losses, and a stock priced for peak earnings gets hit twice — lower earnings and a lower multiple.


Price & moving averages 12 months · 50 & 200-day averages · 52-week range

-1445221,1871,8532,519Jul '25Sep '25Nov '25Feb '26Apr '26Jul '2652w hi $2,335Price 1,74550-DMA 1,610200-DMA 70352w lo $41

Solid = price · dashed = 50-day average · dotted = 200-day average · amber = 52-week high/low. Price above both averages is an uptrend.

Bollinger Bands 20-day average ± 2 standard deviations

-1605271,2131,9002,587Jul '25Sep '25Nov '25Feb '26Apr '26Jul '2620-day avg 1,952Price 1,745

The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.

RSI (14) momentum gauge · 0–100

705030Jul '25Sep '25Nov '25Feb '26Apr '26Jul '26RSI 46.8

Above 70 (red band) = overbought, below 30 (green band) = oversold. Currently 47.

MACD 12 / 26 / 9 · trend & momentum

0Jul '25Sep '25Nov '25Feb '26Apr '26Jul '26signal 165.9MACD 127.4

Blue crossing above amber (bars flip green) = momentum turning up; below (bars red) = turning down. Bar height = the size of that gap.

Relative performance vs S&P 500 & its sector (XLK (sector)), set to 100 a year ago

-3091,1312,5714,0105,450Jul '25Sep '25Nov '25Feb '26Apr '26Jul '26SNDK 3,776XLK (sector) 142S&P 500 120

Solid = SNDK · dashed = S&P 500 · dotted = XLK (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.

Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate

017335066$7BFY24EPS $0$7BFY25EPS $3$20BFY26EEPS $66$45BFY27EEPS $191$51BFY28EEPS $214$59BFY29EEPS $255$12BFY30EEPS $14

Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.

Key stats an RIA wants

Price$1,745.00
Market cap$258B
P/E trailing76×
P/E FY26E / FY27E27× / 9×
EV / Sales19.3×
EV / EBITDA47.0×
Gross margin56.0%
Net margin34.2%
Dividend yield0.00%
Beta4.8812504
52-wk range$41 – $2,335
RSI(14)47
50 / 200-DMA$1,610 / $703
12-mo return+3,781% (SPY +21%)
Street target$1,684 ($450–$3,000)
Analyst grades13 Buy · 2 Hold · 0 Sell
FMP ratingB+
Next earnings2026-08-05

What the experts actually said 1 traceable claims on SNDK · showing the highest-conviction voices

“Memory names look cheap on single-digit P/Es but deserve a discount as the most cyclical stocks; low multiples reflect earnings that swing wildly, not value.”
Compound And Friendsbearishconviction 722026-05-12compound_and_friends-0ioQrA6tYw4:b02c734e9f

Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.

1. What it is

Sandisk Corporation (Nasdaq: SNDK) designs, manufactures and supplies NAND flash storage — solid-state drives (SSDs), embedded memory, removable memory cards, USB devices, and the underlying wafers and components. It was spun back out as a standalone public company (the former Western Digital flash business; David Goeckeler is CEO) and carries ~$5.0B of goodwill on the balance sheet from that heritage. Headquartered in Milpitas, CA; ~12,000 employees; fiscal year ends late June.

NAND flash is a commodity-cyclical business: a handful of producers (Samsung, SK hynix/Solidigm, Micron, Kioxia and Sandisk) supply an undifferentiated bit, so pricing — and therefore margins — swings violently with the supply/demand balance. The current up-cycle is driven by AI-data-center storage demand colliding with disciplined supply.

Revenue mix (FY2025, ended 2025-06-27, from filings):

2. The expert thesis — what the panel says (traceable)

There is no net-bullish expert coverage of Sandisk in the Synthos KB. Total claims = 1, and that one voice is bearish. This verdict is therefore fundamentals- and quant-driven, not conviction-driven — and we say so plainly.

Honest composite note. With breadth of 1 and a net-bearish signal, there is no expert conviction to lean on. We do not manufacture a bull case the KB does not contain. The Street sell-side is constructive (13 Buy / 2 Hold), but that is momentum-following coverage of a name that has already run 38×, and we treat it as context in §6, not as an anchor.

3. Synthos scores & the Bull / Base / Bear cases

The one-glance judgment — three scores, 0–10, each anchored to real metrics (not probabilities we can't honestly calibrate):

Score0–10The read
Downside Risk (higher = riskier)9 · Very HighBeta 4.88, up ~38× in 12 months, priced on peak-cycle EPS; the same business posted a $1.64B net loss in FY25. Balance sheet is fine (net cash on TTM EBITDA) — the risk is the cycle and the multiple, not solvency.
Growth Quality5 · MiddlingTTM margins are spectacular (56% gross, 34% net) but they are a cyclical spike, not a durable trend — FY25 gross margin was 15% and the year was a loss. Real, but not dependable.
Exponential Potential3 · LowThe surge is a NAND up-cycle, not secular acceleration. The estimate curve itself shows a FY30 collapse to ~$11.6B revenue / ~$13.5 EPS after the mid-cycle peak. Commodity memory does not compound.

The three cases (our own scenario model — assumptions shown; each target is a ~12–18-month fair value). We deliberately do not attach probabilities. Memory is a cycle, so we anchor on normalized / peak-cycle earnings times a memory-appropriate (low) multiple, not on trailing GAAP.

CaseKey assumptionsFair value
BullThe AI-storage super-cycle extends through FY27; FY26E EPS of ~$65.6 holds or beats and the market keeps paying a rich-for-memory ~22× peak multiple.~$1,443 (−17%)
Base (our anchor)FY26E EPS ~$65.6 (analyst avg, 10 analysts) is peak-cycle; a memory name earns only a ~15× multiple on peak EPS as the market discounts the coming down-leg.~$984 (−44%)
BearThe cycle rolls over; normalized/mid-cycle EPS reverts toward ~$30 and the multiple stays ~15× on lower earnings — the classic double-hit.~$450 (−74%)

Synthos fair value = the base case, ~$984 (−44%), with the full $450–$1,443 span as the honest range. Note our bull ($1,443) sits below the Street consensus ($1,684) — we think the sell-side is capitalizing peak earnings without a cycle discount, precisely the error the KB's one voice warns against. This is a tracked call — the Forecaster Scorecard grades it once it matures.

4. Exponential Potential

Synthos separates compounders (durable high returns on capital) from exponentials (accelerating multi-baggers-from-here). SNDK is neither — it is a cyclical at the top of its cycle:

Exponential Potential: Low (3/10). Own memory for a cycle trade if you must, sized tiny — not as an exponential compounder. The honest framing is why SNDK sits in the Avoid bucket, not any flagship sleeve.

5. Financials (real numbers — FMP annual/quarterly)

6. Valuation — priced for the top of the cycle

On trailing numbers SNDK is expensive by any measure: 57× EPS, 19.3× EV/sales, 47× EV/EBITDA. The bull leans on the forward: on FY26E EPS of ~$65.6 the P/E is ~27×, and on the (noisy) out-year estimates it screens into single digits. But that is the memory trap the KB voice named: those are peak-cycle earnings, and low forward multiples on peak EPS are not "cheap" — they are the market correctly refusing to capitalize a peak. A through-cycle valuation on normalized earnings (~$30 EPS) at a memory-appropriate ~15× lands near $450–$500; capitalizing the FY26E peak at a still-generous 15× lands near $984 — our base. Street targets (context): consensus $1,684, high $3,000, low $450 — an enormous dispersion that itself signals how cycle-dependent every model here is. Our base sits well below consensus because we apply the cycle discount the Street is not. Not cheap; priced for the peak.

7. Technicals (from the tech block)

8. Moat & competitive position

Sandisk's competitive position is that of a commodity NAND producer — its moat is thin by design. It competes on cost/scale/technology-node against Samsung, SK hynix/Solidigm, Micron and Kioxia in a market that sells an undifferentiated bit. Pricing power exists only when the industry is supply-constrained (now); it evaporates when capacity catches up. There is genuine engineering scale and IP, but nothing that lets Sandisk earn durable excess returns across a full cycle — the FY23–FY25 losses prove it.

Peer set (FMP-supplied, market cap): the listed peers — Credo $45B, Flex $50B, HPE $55B, Jabil $36B, Keysight $54B, MongoDB $29B, Super Micro $18B, Teledyne $30B, Teradyne $58B, Wipro $20B — are a grab-bag of hardware/EMS/semi-cap names, not direct memory comps, so treat them as a loose sector reference only. The economically relevant comparables are the pure-play memory makers (Micron, SK hynix, Kioxia), which historically trade at low single-to-high-single-digit multiples on peak earnings for exactly the cyclicality reason above.

9. Management, capital allocation & guidance

10. Catalysts & what to watch

Thesis tripwires (what would change the call — toward a Watch): a genuine structural shift to under-supplied HBM/enterprise-NAND economics that persists across a full cycle; a re-rating down to a through-cycle valuation near our base (~$984) with the cycle still early; or a durable margin floor well above prior troughs. Absent those, Avoid stands.

11. Key risks

12. Verdict, position sizing & monitoring

Avoid. Sandisk is a competent operator riding a real NAND up-cycle, but at $1,745 — after a ~38× twelve-month run, at peak-cycle margins, with beta 4.88 and officers selling into strength — the stock is capitalizing the top of the cycle. Our through-cycle base fair value is ~$984 (−44%), and even our bull ($1,443) sits below Street consensus ($1,684). The only KB voice covering the name is bearish, and it warns of precisely this trap. There is no fundamentals- or conviction-based reason to buy here.


Provenance & disclosures