Sandisk: From Spin-Off to AI Storage Powerhouse

Sandisk’s transformation has been extraordinary.

After separating from Western Digital in February 2025, the company began operating independently with a focus on NAND flash memory, solid-state drives, and storage technology.

Its fiscal 2026 results demonstrated just how dramatically conditions had changed.

Revenue increased approximately 175%, while the company’s Datacenter business experienced particularly explosive growth.

However, there is an important distinction between revenue growth driven by increased storage shipments and growth driven by higher prices.

In Sandisk’s fiscal fourth quarter, approximately one-third of sequential revenue growth came from higher volumes, while two-thirds came from higher pricing.

That matters because NAND flash is historically a cyclical industry. Strong pricing can generate exceptional profits during shortages, but the reverse can happen when supply catches up.

Sandisk’s challenge is to demonstrate that its financial transformation can survive changing industry conditions.

The Big Change: AI Needs More Than Processing Power

Investors often associate AI infrastructure with advanced GPUs, but computing power is only one part of the equation.

AI systems also require DRAM, high-bandwidth memory, persistent storage, networking, and increasingly sophisticated systems for managing enormous amounts of data.

As inference workloads expand, storage becomes particularly important.

One example is the key-value cache, commonly called the KV cache. This helps AI systems efficiently reuse information during inference. Keeping everything in expensive, high-performance memory is not always economical, creating opportunities for additional storage layers.

Sandisk sees this as a significant market opportunity.

At its August 2026 Investor Day, management projected that enterprise data center flash demand could reach 1.2 zettabytes by 2030.

The company also highlighted its High Bandwidth Flash technology and next-generation NAND products as potential ways to serve increasingly storage-intensive AI applications.

These developments could expand the role of NAND flash beyond traditional enterprise storage.

The opportunity is substantial, but the 2030 demand figure remains a company projection rather than guaranteed future sales.

Sandisk’s New Business Model: Can Contracts Tame the NAND Cycle?

Perhaps Sandisk’s most consequential development is its introduction of multi-year New Business Model agreements, or NBMs.

Historically, NAND manufacturers have been exposed to substantial fluctuations in supply, demand, and selling prices.

Sandisk’s new contracts attempt to improve predictability by securing customer volume commitments, establishing structured pricing arrangements, and providing financial guarantees.

At its 2026 Investor Day, Sandisk reported agreements with eight customers, representing approximately:

• 50% of expected fiscal 2027 NAND bit shipments.
• Two-thirds of expected fiscal 2028 NAND bit shipments.

These commitments could improve capacity planning and reduce exposure to short-term price volatility.

Management’s longer-term financial model for fiscal 2028 through fiscal 2030 targets mid-to-high-teens revenue growth, approximately 80% non-GAAP gross margins, and an adjusted free cash flow margin near 50%.

Those are ambitious targets.

But contracts do not eliminate economic cycles.

If NAND supply increases substantially or customers reduce purchases, the commercial and legal durability of these agreements could become critical.

For investors, the key question is whether the contracts represent a lasting improvement in industry economics or primarily protect Sandisk during today’s unusually favorable environment.

Micron: A Different Way to Invest in AI Memory

Micron Technology offers exposure to several important memory markets.

Unlike Sandisk’s primary focus on NAND flash and solid-state storage, Micron participates in DRAM, high-bandwidth memory, and NAND.

This distinction is important.

High-bandwidth memory, or HBM, is closely associated with advanced AI accelerators. It supplies the bandwidth needed to keep powerful processors fed with data.

DRAM supports computing workloads across servers, PCs, and other devices, while NAND provides persistent storage.

Micron therefore offers a broader memory investment profile, including direct exposure to the high-performance memory required by AI computing systems.

The current AI infrastructure boom has created particularly favorable conditions for HBM and other advanced memory products.

However, Micron also faces substantial capital expenditure requirements, manufacturing competition, and the possibility that today’s elevated memory prices eventually normalize.

Investors should not assume that every dollar of current earnings represents a permanent improvement in profitability.

Sandisk vs. Micron: Comparing the Investment Stories

Both companies benefit from increasing AI infrastructure investment, but their growth drivers are not identical.

Sandisk is particularly interesting because of its effort to transform the financial characteristics of NAND manufacturing.

Micron is particularly interesting because AI accelerators require increasingly sophisticated memory systems, creating opportunities for higher-value products.

For long-term investors, understanding those differences is more useful than simply assuming all memory and storage stocks will perform alike.

What About Western Digital?

Western Digital (NASDAQ: WDC), Sandisk’s former parent, provides another way to participate in the expansion of digital infrastructure.

Following the separation, Western Digital concentrates on hard disk drives.

Although HDDs are sometimes viewed as older technology, they remain important for storing enormous volumes of data economically.

AI workloads can generate substantial amounts of training data, outputs, logs, and archived information.

Not all of that information needs the performance characteristics of flash storage.

For large-scale cloud operators, hard disk drives can remain an attractive solution when storage capacity and cost per terabyte matter more than access speed.

This makes Western Digital a useful comparison, even though its underlying technology and economics differ from those of Sandisk and Micron.

The AI storage opportunity is not necessarily a competition in which one technology replaces every other. Different tiers of memory and storage can grow together.

Valuation: Is Sandisk Already Pricing In the AI Boom?

One of the more interesting questions surrounding Sandisk is whether its share price adequately reflects the company’s growth prospects and risks.

An October 11 analysis by TQP Research used a reverse discounted cash flow model to estimate that Sandisk would need approximately 21.4% annual adjusted free cash flow growth over five years to support the valuation used in that analysis.

The model assumed a 10% discount rate and 2% terminal growth rate.

Those assumptions matter enormously.

Changes in long-term growth, discount rates, or sustainable free cash flow could produce substantially different estimated values.

The analysis also compared projected fiscal 2027 free cash flow with market capitalization, calculating a hypothetical shareholder yield of approximately 12.1% for Sandisk and 12.0% for Micron.

These figures are not dividend yields, guaranteed shareholder returns, or promises that investors will receive cash distributions.

They are valuation comparisons based on projected free cash flow.

Sandisk has announced a substantial share repurchase authorization and intends to return excess cash after investing in its business.

But the amount and timing of actual repurchases will depend on future business conditions and management decisions.

The larger lesson is that strong free cash flow can make an apparently expensive stock look more reasonably valued—but only if that cash flow proves sustainable.

October 29: A Critical Test for Sandisk

Sandisk’s upcoming fiscal first-quarter 2027 earnings announcement, scheduled for October 29, could provide important evidence about the company’s trajectory.

Management previously guided to quarterly revenue between $10.30 billion and $10.80 billion.

For Memoria Byte, five developments deserve particular attention:

1. NAND shipment volumes. Stronger shipment growth would help demonstrate that demand is expanding beyond the benefits of higher selling prices.

2. Datacenter revenue. Continued growth could reinforce the importance of enterprise flash in AI infrastructure.

3. Gross margins. Sustained high margins would support management’s favorable supply-demand outlook, although unusually high profitability also raises questions about future normalization.

4. New Business Model agreements. Additional customers, larger volume commitments, or stronger financial guarantees could improve visibility into future revenue.

5. Free cash flow. Cash generation is essential to evaluating both the company’s valuation and its capacity to return capital to shareholders.

The most encouraging report would combine strong revenue with healthy shipment growth, durable margins, and continued expansion of contractual customer relationships.

A report driven almost entirely by higher prices would still demonstrate favorable current conditions, but it would provide less evidence that the underlying industry has become structurally less cyclical.

Three Risks Investors Should Not Ignore

1. The Memory Cycle Could Return

Memory and storage markets have experienced repeated boom-and-bust periods.

Even if AI creates a permanently larger market, semiconductor manufacturers can still expand supply faster than demand.

That could pressure pricing and profitability.

2. AI Infrastructure Spending Could Slow

Today’s demand depends partly on enormous investments by cloud providers and technology companies.

A slowdown in capital expenditure, changes in AI system architecture, or weaker-than-expected returns on AI investment could affect memory and storage demand.

3. Exceptional Margins May Not Last

High margins attract investment and competition.

Even when demand remains strong, improving manufacturing yields, new production capacity, and technological changes can alter the balance between supply and demand.

Investors should examine normalized earnings potential rather than assuming current profitability will continue indefinitely.

Memoria Byte’s Outlook: A Golden Age, but Not a Risk-Free One

The memory and storage industry is experiencing a remarkable transformation.

AI is increasing the importance of technologies that once received considerably less investor attention than processors and accelerators.

Sandisk offers exposure to NAND flash, enterprise SSDs, and an ambitious contractual business model designed to improve financial predictability.

Micron offers exposure to DRAM, high-bandwidth memory, and NAND, giving it a different set of opportunities within the AI ecosystem.

Western Digital demonstrates that even traditional hard disk drives can play an important role in the expanding digital economy.

My conclusion is that AI could support a prolonged period of growth across multiple layers of the memory and storage industry.

But a growing market does not automatically guarantee attractive stock returns.

Valuation, capital spending, pricing discipline, competitive dynamics, and the durability of free cash flow remain essential.

For Sandisk, the October 29 earnings report will provide the next major test.

For Micron, investors should continue evaluating how AI-driven demand translates into sustainable profits across its memory portfolio.

The AI revolution needs processors to think, memory to work, and storage to remember.

And that’s why these companies deserve a place on the radar of investors following the next generation of computing infrastructure.

— Memoria Byte, Memory & Data Storage Stocks Analyst

Alluring Analysts | AnalyzeStocks.com

Disclaimer: This article is provided for informational, educational, and entertainment purposes only. It is not personalized investment advice or a recommendation to buy or sell any security. Financial projections, valuation models, and forward-looking statements are subject to uncertainty. Investors should conduct their own research and consider their financial circumstances and risk tolerance before making investment decisions.