Could Databricks at $100B Be … Cheap? Why Growing 2x Faster Than Snowflake Might Make It the Better Buy

In most markets, a $100 billion valuation is the point where people start whispering “bubble” and refreshing their portfolio apps every five minutes.
But in the world of data and AI infrastructure, the question around Databricks is strangely different:
could this thing actually be cheap at $100B?

With Databricks reportedly growing roughly twice as fast as Snowflake at a similar revenue scale, a lot of SaaS investors and operators are quietly
asking themselves whether the real risk is not owning enough of the category winner, rather than overpaying by a few billion on the way up.

In this deep dive, we’ll unpack why Databricks’ $100B valuation may be more rational than it first appears, how its growth compares to Snowflake’s,
and what signals to watch if you’re trying to decide which data giant looks like the better long-term bet.

Where Databricks and Snowflake Stand Today

The headline numbers

Let’s set the stage with a few key datapoints:

• Databricks: The company recently announced it is raising a Series K round at a valuation above
  $100 billion, off the back of a $4 billion+ revenue run rate growing at 50%+ year-over-year and
  more than $1 billion in AI-specific revenue. The business is also free-cash-flow positive and boasts net revenue retention
  north of 140%.
• Snowflake: Snowflake, the incumbent public data cloud leader, is generating roughly
  $3.6–$4.0 billion in annual revenue, with product revenue growth in the high-20s to low-30s percentage range and strong
  free cash flow margins.

In other words, Snowflake is slightly bigger on absolute revenue today and already public, while Databricks is
growing roughly 2x faster at nearly the same scale, with private markets rewarding it with that eye-popping $100B price tag.

Same destination, different roads

Databricks and Snowflake often get lumped together as if they’re Coke vs. Pepsi. In reality, they’ve grown up with
different starting points and strengths:

• Snowflake started as a cloud data warehouse with a heavy emphasis on SQL analytics, BI, and reporting.
  Its “Data Cloud” positioning aims to be the universal layer where organizations centralize and share data.
• Databricks began as a data lake and AI/ML platform built on Apache Spark, then evolved into a
  “lakehouse” that blends data warehousing, data engineering, and AI workloads on top of open formats.

Over the last few years, Snowflake has moved aggressively up into AI and application workloads, while Databricks has moved
down into classic warehousing and SQL analytics. The overlap is now huge. To customers, it increasingly feels like a
platform duel for all things data and AI.

How Can a $100B Startup Be “Cheap”?

Calling a $100B valuation “cheap” sounds like calling a private jet “fuel efficient.” But in long-duration software and
infrastructure, the price tag only makes sense when you zoom out to the next decade.

Looking at revenue multiples in context

Suppose Databricks is at a ~$4B revenue run rate growing 50%+ year over year. A $100B valuation implies a
revenue multiple in the ballpark of 25x forward revenue (depending on exactly how you define run rate and forward look).
That’s a nosebleed multiple in most sectorsbut not unprecedented for category-defining infrastructure names with huge TAMs
in the middle of a platform shift (think early days of AWS-type businesses or the top-tier cybersecurity leaders).

Now compare that to Snowflake. With growth now in the high-20% range and revenue already in the multi-billion territory,
public markets have gradually compressed Snowflake’s multiple versus its early, hyper-hype years.
Snowflake still trades rich relative to traditional SaaS, but less so relative to its own history.

If Databricks is truly growing about twice as fast at nearly the same scale, you can make an argument that:

• Either Databricks deserves a meaningfully higher revenue multiple than Snowflake today,
• Or Snowflake is actually undervalued and the whole data/AI complex should be repriced upward.

In practice, investors tend to assume the truth lies somewhere in the middle. Databricks gets the growth premium;
Snowflake gets the public-market liquidity and stability premium. The question then becomes: who has the better
long-term compounding engine?

The rule of 40 (and then some)

Classic SaaS investors love the Rule of 40: growth rate + free-cash-flow margin > 40%. Databricks, at 50%+ growth with
positive free cash flow, blows past that threshold. Snowflake, with high-20s growth and strong free cash flow, also screens
extremely well.

The difference is that at Databricks’ growth pace, compounding works like a monster. Add 50%+ a year for a few years
and your ARR doubles faster than you can spin up another cluster. If Databricks can sustain even 35–40% growth at scale for
several more years while maintaining healthy margins, the $100B headline starts to look more like a waypoint than a peak.

Why Databricks Is Growing 2x Faster Than Snowflake

1. Riding the AI platform wave

The first and most obvious driver: AI demand. Databricks has positioned itself at the intersection of data,
machine learning, and generative AI with a “data intelligence platform” story that resonates deeply with enterprises
trying to activate their existing data for AI use cases.

Key tailwinds include:

• Enterprises need massive, flexible data pipelines to feed LLMs and AI agents. Databricks’ lakehouse and
  open-format approach (Delta Lake, Parquet, etc.) is very appealing for this.
• Databricks has leaned heavily into AI-native workloadsfrom model training and fine-tuning to vector search,
  feature stores, and agent frameworksmaking it feel like the natural “home base” for AI development teams.
• A growing chunk of its run rate (over $1B) is directly tied to AI products and use cases, which tend to be high-consumption,
  high-expansion workloads.

Snowflake is also investing aggressively in AI (Snowflake Cortex, Snowpark, acquisitions in the ML and AI stack),
but it started from a more traditional analytics base and is still in the process of repositioning itself in the minds of many customers
as the “AI app platform,” not just the data warehouse.

2. Open formats and ecosystem gravity

Databricks’ embrace of open formats and open source (Spark, Delta Lake, MLflow, and now open table formats like Iceberg interoperability)
creates a strong ecosystem pull:

• Customers like the optionality and lack of lock-in; it feels safer building critical AI workloads on open formats
  and standards.
• Partners (ISVs, SI firms, consultants) find it easier to integrate and build around a platform that doesn’t require deep
  proprietary magic for every interaction.
• Talent is easier to hire when the skills are based on widely adopted tools and frameworks rather than a single closed platform.

Snowflake’s moat has historically been its managed, tightly integrated, proprietary cloud data stack, which delivers
exceptional performance and simplicity for analytics but can feel more closed relative to the Databricks ecosystem.
As AI becomes more experimental and fast-moving, that openness can translate directly into faster platform adoption.

3. Consumption, net retention, and big-ticket customers

Both Databricks and Snowflake run on consumption-based pricing, which creates a powerful expansion engine.
But Databricks has been especially strong in three areas:

• Very large customers: Hundreds of customers reportedly spend over $1M annually, with many marquee enterprises
  running critical AI and data workloads.
• Net revenue retention >140%: That’s elite even in the upper echelons of infrastructure SaaS.
  It means existing customers keep growing their usage very rapidly.
• Cross-product expansion: Customers might start with data engineering or ETL, then expand into SQL analytics,
  streaming, and finally AI and agentseach new workload layer adds more compute and storage consumption.

Snowflake’s net retention is also strong (mid-120s), but that extra 10–20 points at Databricks multiplied across
multi-billion-dollar revenue bases is exactly what drives the “growing 2x faster” headline.

4. Product breadth vs. depth trade-offs

Databricks is playing a slightly different game:
one platform for data engineering, BI, and AI, all built around the lakehouse.
That breadth can be overwhelming at times, but it also means:

• Fewer data hops between systems (warehouse → lake → ML platform → vector DB, etc.).
• Potentially lower total cost of ownership for customers willing to standardize.
• A simpler story for executives: “Everything data + AI, one platform, one commercial relationship.”

Snowflake, while expanding quickly, has historically led with depth in analytical workloads and is adding
more pieces to cover AI and application development. The market is effectively voting, with budgets, on which
story feels more compelling.

Why Databricks Might Be the Better Buy (In Theory)

Let’s be clear: whether either stockor eventual stock, in Databricks’ caseis a good personal investment
depends on your situation, risk tolerance, and time horizon. Nothing here is financial advice.

That said, from a purely strategic and growth-investor lens, there are several reasons Databricks might look like the
more attractive long-term compounder at $100B than Snowflake at a similar or even lower market cap.

1. Growth vs. scale curve

At the same rough multi-billion revenue scale:

• Snowflake is growing around 25–30% a year.
• Databricks is growing ~50%+ a year.

If Databricks can maintain even a modest lead on growth over the next 3–5 years, its absolute revenue can pull materially
aheadeven if Snowflake remains a strong performer. Revenue leadership in a category often compounds into ecosystem leadership
(more partners, more talent, more integrations, more “default choice” behavior).

2. AI-native positioning

The market is increasingly assigning premium multiples to platforms viewed as AI-native, not just AI-enabled.
Databricks’ brand, product history, and customer references tilt heavily in that direction.
It’s easier to add more classic analytics onto an AI-native platform than to convince the world that your analytics platform
is suddenly the best place to do all their AI.

3. Open ecosystem and optionality

The bet Databricks is makingthat the future will be built on open formats, open integration, and multi-cloud flexibilitygives it
a lot of strategic optionality:

• It can partner broadly across clouds and vendors without threatening its core value proposition.
• It can lean into whichever AI stacks, chips, and deployment patterns win over the next decade.
• It can afford for some product bets to fail as long as the core lakehouse and AI platform keep compounding.

Snowflake’s proprietary layer has been a huge plus for performance and simplicity, but it may have to work harder
to reposition around openness and interoperability to avoid friction with the broader AI ecosystem.

4. The private-to-public re-rating potential

One of the subtle reasons some investors get excited about a $100B Databricks is the future re-rating story.
If the company goes public and continues to execute, public markets could eventually reward it with a valuation significantly
higher than its last private roundespecially if:

• AI spending accelerates and Databricks is seen as a central beneficiary.
• Growth stays above 35–40% at multi-billion scale.
• Free cash flow remains robust, proving the business model is structurally efficient, not just top-line flashy.

That’s the essence of the “cheap at $100B” argument: not that the price is low in absolute terms, but that
the long-term compounding could outpace even aggressive expectations.

Risks That Could Break the Bull Case

Of course, there’s a non-zero chance all of this ages like a 2021 SPAC deck. A few things to keep in mind:

• Platform sprawl risk: Databricks’ breadth is a strength, but also a UX and go-to-market challenge.
  If customers feel overwhelmed or under-served on core use cases (like straightforward BI), they may hedge with Snowflake
  or other specialized tools.
• AI hype vs. durable demand: If AI spending proves more cyclical than expected, or if many proof-of-concepts
  never go into production, consumption could slow, pressuring that 50%+ growth story.
• Competitive responses: Snowflake, hyperscalers (AWS, Azure, GCP), and a long tail of data/AI startups are
  not quietly watching Databricks sprint away. Aggressive pricing, bundling, and innovation could nibble at Databricks’ edge.
• Regulation and data residency: As AI and data privacy regulations tighten globally, the cost and complexity
  of compliance could rise, favoring players with the deepest compliance and go-to-market muscle.

In other words, Databricks still has to earn its $100B valuation every quarter for years. There’s no free lunch,
not even in Silicon Valley.

So… Is Databricks Really the Better Buy Than Snowflake?

Framed purely as a strategic bet on the future of data and AI, you can make a strong case that:

• Databricks is the higher-growth, AI-native, open-ecosystem platform with enormous upside if it keeps
  executing and AI spending keeps compounding.
• Snowflake is the more mature, analytics-centric, public-market leader with strong free cash flow,
  lower growth, and a slightly more conservative risk profile.

Whether Databricks at $100B is “cheap” comes down to a simple but brutal question:
Do you believe it can 5–10x from here over a decade?
If the answer is yes, then $100B is just another milestone. If the answer is no, the headline number already bakes in a
lot of the good news.

The only thing that seems increasingly clear is that this is not a winner-takes-all market. Many enterprises will run both
Databricks and Snowflake side by side for years, with workloads gravitating toward whichever platform best fits their mix of
data engineering, analytics, and AI. The real “better buy” might ultimately be whichever one you understand well enough to hold
through the inevitable volatility.

Experiences from the Trenches: How Operators and Investors Think About Databricks vs. Snowflake

It’s one thing to stare at growth rates and valuations from afar. It’s another to live through them from the inside.
Here are some composite, anonymized experiences and patterns that operators and investors often describe when they talk about
Databricks versus Snowflake.

From the VP of Data: “We ended up with both, and that was fine.”

One common story goes like this: a company starts with Snowflake as the default warehouse. The BI team loves it, finance loves it,
and dashboards appear faster than you can schedule another standup. But as the data science and ML teams ramp up, they begin pushing
for something better suited for large-scale feature engineering, notebook-driven exploration, and deep model pipelines.

That’s when Databricks sneaks infirst as a small, experimental cluster for a fraud-detection or personalization use case,
then gradually as the default environment for all things ML and AI. Before long, the company is paying meaningful money to both platforms,
but no one wants to rip either out because:

• Snowflake is still fantastic for standardized reporting and self-service analytics.
• Databricks is where the higher-value AI and advanced analytics workloads live.

From that VP’s perspective, Databricks looks like the “growth engine” of future data initiatives, while Snowflake looks
like the reliable “system of record” for analytics. That mental model often mirrors how investors think about the two companies.

From the CFO: “The Databricks bill scared me, then I realized why.”

Another recurring experience is the first time a CFO really looks at the Databricks bill. Because the platform is so
tightly coupled to compute-heavy workloads like model training, streaming, and large-scale ETL, the consumption curve can be steep.

At first, that might feel alarminguntil you map the cost back to revenue-generating use cases. When the FP&A team
connects the dots between Databricks consumption and things like better ad targeting, lower fraud losses, or higher conversion rates,
the spend becomes easier to justify.

By contrast, Snowflake spend often feels easier to predict: dashboards, scheduled queries, batch analytics. It’s not that Snowflake
can’t power revenueof course it canbut the direct line between “more Databricks” and “more AI-driven business outcomes” can feel
more tangible in the current cycle.

From the investor: “I’m not betting on features, I’m betting on gravity.”

Experienced SaaS and infrastructure investors rarely obsess over a single product announcement. Instead, they talk about
gravitywhich platform is attracting:

• The best customers with the biggest, hairiest data problems.
• The strongest partners and ecosystem players.
• The most ambitious developers who want to build the next wave of AI-native apps.

In conversations like that, Databricks comes up again and again as the place where “the weird, cutting-edge AI experiments”
are happening at scale. Snowflake is praised for its discipline, financial quality, and strong executionbut Databricks is
often described as the higher-beta, higher-upside platform that might define the next generation of AI data infrastructure.

That doesn’t mean every investor loads up on Databricks exposure at any price. But it explains why some are surprisingly relaxed
about a $100B headline. If they believe the gravitational pull is shifting toward Databricks as the default AI data platform,
the real risk is missing the compounding more than paying a full price upfront.

From the startup founder: “We built on Databricks because that’s where our customers live.”

Finally, there’s the early-stage founder who has to choose: build deeply on top of Databricks, Snowflake, or try to stay neutral.

If their target buyer is a data science or AI team, they often find Databricks already deeply embedded. Using Databricks as the
“home base” for their integration gives them:

• Closer alignment with AI-heavy workflows.
• Access to customers who are already experimenting with cutting-edge ML.
• A natural story: “We extend what you’re already doing in Databricks.”

That kind of grassroots founder behavior doesn’t show up directly in quarterly earningsbut over time, it reinforces the notion that
Databricks is where the future of AI data work is being prototyped and scaled.

Taken together, these experiences explain why the idea of Databricks at $100B “still being cheap” doesn’t sound quite as crazy
once you’ve lived in the ecosystem. It doesn’t mean the outcome is guaranteed, and it doesn’t mean Snowflake is standing still.
It does mean that if Databricks continues to be where the most ambitious AI and data projects are born, the compounding may still
be in its early chapterseven at a valuation that already looks larger than many countries’ GDPs.