Sandisk (SNDK) Deep Dive: Why We Own It

Why We Own Sandisk (SNDK)

Sandisk is the rarest combination in our universe right now: hyper-growth that the data still calls undervalued. Over the TTM, the company has grown revenue +79.7% — over 5x the GICS Information Technology sector median (15.2%, n=71). Net margin sits at 34.2%, nearly twice the sector. Forward P/E is 23.4x against roughly +65% expected forward EPS growth — a PEG ratio close to 0.4.

The company is a current Veloris Capital portfolio holding. SNDK was selected systematically through our three-pillar process — stock universe screen, equal-weight optimizer, risk overlay — not because of the price chart, but because the underlying data signed off on every step.

At a Glance

The benchmark used is the broad GICS Information Technology sector: top 71 U.S.-listed names by market cap, $500M minimum cap, computed from primary fundamentals data. The U.S.-listed direct peer universe (Micron, Seagate, Western Digital) is too small (n=3) to call a defensible median, so this article uses the broad sector as the only formal benchmark. Where storage and memory peers add useful cycle context, they appear inline.

A Memory Cycle Inflection With a Pure-Play Beneficiary

The full storage and memory complex is in a strong cyclical upturn. Within that cohort, SNDK's revenue growth (+79.7% TTM) sits roughly in line with Micron's (+85.5%) and well ahead of Seagate's (+28.9%) and Western Digital's (+9.1%) — and on EBITDA margin (42.7%) it leads all three. Against the broad sector median (revenue +15.2%, EBITDA margin 28.9%), Sandisk is operating in a different growth and margin regime entirely.

Three factors are driving the divergence within the cycle: (1) Hyperscaler AI-training workloads shifting to high-capacity SSDs as model checkpoints and dataset staging stress storage tiers; (2) post-spinoff customer agreement rewrites that lock in multi-year volume at predictable pricing; (3) limited near-term NAND supply elasticity since competitor capacity additions take 18–24 months to ramp. The result is a margin and growth profile that screens differently from the rest of the storage cohort.

Analyst revision behaviour confirms the asymmetry. Twenty sell-side analysts cover the FY+1 forecast period. In the last 30 days, 9 raised their EPS estimates and zero cut. In the last 7 days, 3 more raised and zero cut. That is a unanimous bull pattern — no analyst has reduced numbers in over a month even after the post-spinoff re-rating. Either the entire street is anchoring to the wrong trajectory together, or the underlying fundamentals are running ahead of expectations.

The Business — A NAND Pure-Play Inside a Concentrated Industry

Sandisk became a publicly traded company in February 2025 after separating from Western Digital. The legacy entity kept the hard-disk-drive business (HDDs); the spinoff kept the NAND flash side. SNDK now sells solid-state drives for desktops, notebooks, gaming consoles, and datacenter servers; embedded storage modules for mobile phones, tablets, and automotive applications; and removable cards, USB drives, and flash components used across consumer and industrial markets.

NAND is a four-player industry globally — Samsung, SK Hynix, Kioxia, and the SNDK/Kioxia joint venture combined account for the overwhelming majority of world output. Sandisk's manufacturing footprint runs through that joint venture, which is the world's #2 NAND production base. That structure is the most important fact about the business.

“Why Sandisk, not "any NAND maker"? Because the Kioxia joint venture provides access to fab capacity at the world's #2 producer without bearing the full capex of building it — a structurally lower-cost capacity ramp than competitors who have to fund their own gigafactories.”

Institutional-Grade Profitability

SNDK's 34.2% TTM net margin runs at roughly 1.9x the broad sector median (18.4%). For storage and memory cycle context: Micron (41.5%) currently runs higher because of DRAM exposure that's in a different cycle phase, and Western Digital's post-spinoff TTM (55.1%) reflects accounting gains tied to the Sandisk separation itself; once those normalize the gap narrows materially. Seagate, by contrast, runs at 21.6%.

EBITDA margin is the cleaner cycle-adjusted read. SNDK's 42.7% runs roughly 14 points above the broad sector median (28.9%) and ahead of every U.S.-listed storage and memory peer we track. This is best-in-class operating leverage during the upcycle — the kind of structural number that reveals whether a business has a cost moat or just a tailwind.

The balance sheet adds an institutional check: $3.74B in cash and equivalents, $13.78B in book equity, and a modest debt load. The company is approaching this cycle with a stronger balance sheet than the industry carried into 2017's peak — which is what creates room for the long-term customer agreements that have changed the revenue model.

The Earnings Trajectory

The most striking single data point in this name is what just happened to the earnings line.

Trailing-twelve-month EPS is $31.12. Forward consensus from the 20-analyst panel implies ~$54 over the coming 12 months — another ~+73% from the current run-rate. With 9 upward revisions and zero cuts in the last 30 days, the consensus is moving toward the company, not the other way around.

The strategic shift toward multi-year customer agreements is what underwrites the durability of this trajectory. Long-term contracts trade some upside in spot NAND pricing for stability across the cycle — the trade-off that institutional copiers should care about, because it reduces the catastrophic-quarter risk that has historically defined memory investing.

Valuation — Premium That Is Mostly Earned

SNDK trades at a 23.4x Forward P/E against roughly +65% expected forward EPS growth. That is a PEG ratio of approximately 0.36 — a multiple genuinely cheap relative to the growth trajectory. The trailing P/E of 40.5x looks expensive on its face, but trailing earnings have not yet absorbed the explosive Q1 2026 quarter; six months from now the trailing multiple will compress materially even at an unchanged price.

P/B of 12.75 is a modest premium to the broad sector (9.74) — about 30% — which is exceptionally narrow given SNDK is growing 5x the sector. Among storage peers individually: Micron sits at 8.44, Western Digital at 15.11, and Seagate at an outlier 148.86 (years of buybacks have compressed book equity to a sliver). Sandisk's P/B is therefore not the data point that signals expensive — the apparent stretch is in the trailing P/S (~14x), and that is a snapshot multiple that has not yet caught up to the run-rate revenue.

The expensive-looking multiple most often cited is the trailing P/S (~14x). It is a snapshot that has not caught up to the run-rate revenue. If the cycle peaks earlier than the long-term contracts suggest, all of these multiples re-rate. Our risk overlay (Pillar 3) is what governs that exposure — it is the entire point of running the strategy with a daily-monitored equity dial rather than a buy-and-hold posture.

How Sandisk Passed Our Three-Pillar Process

Position selection is the output of a systematic process, not a discretionary thesis. SNDK cleared each of the three pillars on its own merits, evaluated against both peer-set medians and our internal thresholds.

Pillar 1 — Stock Universe Filter

The first screen tests revenue acceleration, profitability, balance sheet quality, and analyst-revision momentum against absolute thresholds we apply to every name. SNDK cleared the revenue growth threshold by a wide margin (79.7% TTM vs sector 15.2%), the EBITDA margin threshold (42.7% vs sector 28.9%), the cash-position threshold (>$3B liquid), and the revision-momentum threshold (9-up / 0-down over 30 days). Read more about what the universe filter does in our portfolio construction post.

Pillar 2 — Optimizer (Selection, Not Sizing)

The optimizer decides which 15–30 names from the qualified universe enter the portfolio — it does not decide weight. Every position is equal-weighted. SNDK currently sits in the live portfolio at 6.67% weight alongside fourteen other names (1/15 = 6.67%). It was selected because of forward-growth + revision momentum + balance-sheet quality together — not because of price momentum alone.

Pillar 3 — Risk Overlay

The risk overlay tracks 23 named indicators across volatility, credit, breadth, sentiment, and macro stress, and dictates the strategy's aggregate equity exposure — currently in cruising mode. SNDK has high Beta, which means its position-level volatility is real. The overlay does not solve high-beta names by under-weighting them inside the portfolio (we equal-weight). It solves them by managing the dial of total equity exposure when overall market stress rises. For a worked example of how that dial moved during a real stress event, see how the overlay managed March 2026.

Catalysts on the Radar

• Calendar Q2 2026 earnings — the next quarterly read on hyperscaler AI capex translation into NAND orders.
• QLC enterprise SSD ramp into AI training and inference clusters — the high-capacity tier where SNDK has differentiated product positioning.
• Hyperscaler capex updates from Q1 and Q2 2026 reporting season — direct read on whether AI infrastructure spend is sustaining or rolling over.
• NAND supply expansion from competitors expected in 2027 — the principal medium-term margin risk. Long-term customer agreements partially insulate against spot pricing volatility.
• Multi-year customer agreement renewals — incremental visibility on the revenue stability story.

“Discipline compounds. We do not own Sandisk because of the price chart — we own it because the data screened cleanly: revenue growth 5x the broader sector, profitability that has structurally re-rated, analyst revision momentum running unanimous up, and a forward multiple that is genuinely cheap relative to the trajectory.”

For context on how this name fits the broader portfolio, see our latest monthly review. For the construction logic behind the equal-weighted 15–30 name structure, see why we hold 15–30 stocks. For the broader story of how the strategy was built and where it stands, see our Q1 2026 review.

*Sandisk Corporation (SNDK) is a current Veloris Capital portfolio holding as of the publication date. This analysis is provided for educational purposes and does not constitute investment advice. Past performance is not an indication of future results. Your capital is at risk.*

Weekly investment newsletter, every Sunday — positioning, performance vs SPY/QQQ, and what we're watching next. No spam.