To The Moon

More Than A Fancy Filing Cabinet

The market keeps asking whether Snowflake can defend its data warehouse. I think that's a bit like valuing an airport by its baggage claim.

The real battle is taking place elsewhere.

Snowflake is attempting one of the most ambitious transitions in enterprise software: evolving from a platform that stores information into a platform where autonomous software agents execute work. If management succeeds, the company's future will be determined less by how much data customers store and more by how many decisions, workflows and business processes run through its ecosystem.

That distinction matters because storage is steadily becoming commoditised. Execution is not.

This is why I believe the most important investment question facing Snowflake today is no longer data storage but enterprise AI execution density.

The real prize may be controlling traffic, not storage

When Every Query Becomes A Toll Booth

The rise of enterprise AI is changing Snowflake’s economics in ways many investors have yet to fully appreciate.

Historically, customers used Snowflake for scheduled analytics workloads. Queries were predictable, repeatable and relatively efficient.

AI changes that equation.

Autonomous agents, reasoning systems and workflow orchestration tools consume significantly more compute resources than traditional database activity. A single AI-driven procurement process may trigger dozens of actions, validations and decision points that previously required human involvement.

That creates an important shift.

Snowflake no longer needs customers simply to store more data. It needs them to execute more workflows.

If AI adoption scales as management hopes, revenue growth could increasingly be driven by compute intensity rather than storage expansion.

This is where I believe the market may be underestimating the opportunity.

Many investors still focus on customer growth, retention rates and storage consumption. The metric I increasingly care about is how much computational activity each customer generates.

A platform embedded in daily enterprise decision-making can monetise thousands of small actions. The economics begin to resemble a digital toll road rather than a software subscription.

And toll roads tend to collect more fees when traffic increases.

The Numbers Behind The Narrative

The financial results suggest a company navigating a major architectural transition.

Trailing twelve-month revenue has reached $5.03 billion, while quarterly revenue growth remains impressive at 33.5% year-on-year. At an $83 billion market capitalisation, $Snowflake(SNOW)$ is already operating at meaningful scale, making that growth rate particularly notable.

Yet profitability remains elusive.

The company reported a net loss of approximately $1.2 billion over the trailing twelve months, with profit margins of -23.8% and operating margins of -22.2%.

Viewed in isolation, those figures look unattractive.

However, Snowflake generated approximately $1.24 billion in operating cash flow and reported roughly $1.74 billion in levered free cash flow over the same period. While investors should always scrutinise cash flow adjustments carefully, the contrast highlights an important reality: Snowflake's underlying cash-generating capacity appears considerably stronger than its income statement suggests.

That distinction matters because the company is funding much of its AI expansion internally rather than relying heavily on external capital.

The balance sheet also remains reasonably resilient. Snowflake holds $2.95 billion in cash against total debt of $2.77 billion, giving management flexibility as it continues investing in AI infrastructure and product development.

Valuation, however, remains demanding.

The shares trade at approximately 16 times sales and more than 120 times forward earnings. Investors are clearly paying for future potential rather than current profitability.

Premium valuations rarely forgive execution mistakes.

The Cost Of Thinking

The decisive investment question is not whether AI increases demand.

It almost certainly will.

The real question is whether Snowflake can increase compute intensity faster than infrastructure costs rise.

This is the metric that deserves far more attention than it currently receives.

Generative AI workloads carry fundamentally different economics from traditional data warehousing. Every reasoning task, inference request and autonomous workflow consumes computational resources that must ultimately be funded.

Many investors assume AI automatically improves profitability. In reality, AI can become an expensive habit if revenue fails to outpace infrastructure costs.

The market is testing Snowflake's next operating range

The framework I use is relatively straightforward.

If Snowflake can continue delivering revenue growth comfortably above 25% while gradually stabilising gross margins and expanding operating cash flow, then the AI transition is likely creating economic leverage.

However, if AI-related revenue growth remains robust while margins continue deteriorating and cash flow growth stagnates, management may simply be purchasing growth through increasingly costly compute consumption.

In short, I am less interested in how many AI products Snowflake launches and more interested in whether those products generate profitable computational density.

That scoreboard will reveal whether Snowflake is building a better business or merely a busier one.

One overlooked risk for both bulls and bears is that AI may ultimately make enterprise software revenue less predictable. Traditional SaaS businesses monetised users and subscriptions; Snowflake is increasingly monetising activity. That could create greater revenue upside during periods of intense AI adoption, but potentially more volatility when enterprise workloads slow.

The Iceberg Beneath The Iceberg

The market's view of Apache Iceberg may be one of the most interesting misconceptions surrounding Snowflake today.

Conventional wisdom treats open data formats as a threat because they weaken traditional storage lock-in. Under this view, enterprises gain greater ownership and portability of their data, reducing dependence on Snowflake’s proprietary architecture.

That risk is real.

Yet I believe many investors are focusing on the wrong layer of the technology stack.

Storage is becoming commoditised regardless of who wins. The larger prize sits above storage.

Open architectures allow enterprises to retain ownership of data while simultaneously accessing multiple compute environments. Many organisations are comfortable adopting an orchestration layer even when they are reluctant to commit entirely to a proprietary storage platform.

This creates what I view as the Iceberg Paradox.

The very technology that appears to weaken Snowflake’s traditional moat could expand its future opportunity.

If Snowflake Intelligence becomes the preferred environment where enterprise AI agents execute tasks, then open formats may increase adoption rather than reduce it. Customers gain flexibility over data ownership while Snowflake captures value through workflow execution and compute consumption.

Conviction often hides beneath price volatility

This is also where the competitive battle becomes more interesting.

Most investors assume $Snowflake(SNOW)$ competes primarily against cloud hyperscalers. I am not convinced.

The real competition may be over who controls enterprise orchestration. Cloud providers own infrastructure. Snowflake is attempting to own coordination across increasingly fragmented data environments.

Historically, coordination layers have often proved more valuable than the underlying infrastructure. Operating systems became more valuable than hardware manufacturers. Payment networks became more valuable than many of the physical banking assets beneath them.

Snowflake is attempting to position itself as the coordination layer for enterprise AI.

That does not guarantee success, but it fundamentally changes the nature of the opportunity.

Why Gravity Still Wins

One insight that receives surprisingly little attention is the role of proximity.

Many investors assume enterprise AI agents will operate seamlessly across multiple environments. In practice, corporate data remains fragmented, regulated and highly sensitive.

As AI agents become more sophisticated, moving data around becomes less attractive than moving compute towards the data itself.

This creates a gravitational pull that may ultimately benefit Snowflake.

If Snowflake Intelligence becomes the preferred runtime environment for enterprise workflows, the company could capture value from every automated decision and process executed within its ecosystem.

That is a much larger opportunity than storing information.

It is an opportunity to become part of the enterprise operating layer itself.

Most investors see baggage. I see the tower

Crown, Trap Or Something In Between?

Snowflake remains a polarising investment because both bulls and bears possess legitimate arguments.

The bears are correct that valuation remains rich, margins remain under pressure and open architectures are reshaping the competitive landscape.

The bulls are equally correct that revenue growth remains strong, enterprise adoption continues expanding and AI-driven compute demand could unlock a significantly larger addressable market.

Personally, I think the market is still framing the debate incorrectly.

The future of Snowflake is unlikely to be decided by who stores the most data. It will be decided by who executes the most enterprise work.

If Snowflake successfully transforms itself into the orchestration layer for autonomous business processes, today's concerns about storage commoditisation may ultimately prove secondary.

That outcome is far from guaranteed.

Yet when I look at Snowflake today, I do not see a company fighting to protect a database franchise. I see a company attempting to position itself at the centre of enterprise decision-making.

Most investors are focused on the baggage claim. I am focused on the control tower.

And if management succeeds, Snowflake may ultimately be remembered not as a data company that embraced AI, but as an AI execution platform that happened to start with data.

@TigerStars @Daily_Discussion @Tiger_comments @Tiger_SG @Tiger_Earnings @TigerClub @TigerWire