No matter how you slice the data, Berkshire Hathaway (NYSE:BRK-A)(NYSE:BRK-B) CEO Warren Buffett is one of the greatest investors and leaders of our generation. Since 1965, Berkshire's stock has averaged a compound annual return of 20.3%. That's more than double the annual average return of the broad-based S&P 500, which, inclusive of dividends, returned 10% annually over the same stretch. In total, Buffett has seen his company's stock rise by more than 2,700,000%, and he's created over $400 billion in value for shareholders.

Given the Oracle of Omaha's resounding success on the investing front, it should come as little surprise that Wall Street and retail investors wait on the edge of their seats to see what he's been buying and selling. Occasionally, these trades can lead to investors riding Buffett's coattails to big gains.

Berkshire Hathaway CEO Warren Buffett at his company's annual shareholder meeting.

Berkshire Hathaway CEO Warren Buffett. Image source: The Motley Fool.

Say what -- Buffett just bought a cloud IPO?

But every once in a while, Buffett does something completely out of the ordinary.

Earlier this week, an amended S-1 filing with the Securities and Exchange Commission notes that Berkshire Hathaway agreed to purchase $250 million worth of Class A shares of software-as-a-service (SaaS) cloud stock Snowflake in a private placement, prior to its initial public offering (IPO). The move will allow Buffett's company to scoop up an estimated 3,125,000 shares at an assumed listing price of $80. Salesforce.com is also taking a $250 million private stake.

The S-1 also notes that Berkshire Hathaway has agreed to purchase 4,042,043 shares of Class A stock from one of the company's stockholders in a secondary offering. Buffett's company will be able to make this purchase at the same cost as the IPO price. This means Berkshire would own up to 7,167,043 shares, with an initial investment of approximately $573.4 million (again, based on the assumed $80 IPO list price). 

A cloud in the middle of a data center that's connected to multiple wireless devices.

Image source: Getty Images.

This Snowflake is red-hot

What makes this such a jaw-dropping move is the fact that Snowflake is arguably the most anticipated technology/cloud/SaaS IPO of 2020, if not of the past couple of years.

Over the past eight quarters, Snowflake's revenue has surged from $29 million to $133 million (that's sequential quarterly growth of 24.3%). It also sports 56 customers spending in excess of $1 million annually (up from 22 in the prior-year period), and a ridiculously impressive net revenue retention rate of 158%, suggesting that its existing clients are spending considerably more as they scale their businesses. 

Snowflake is a cloud-native platform, meaning that it'll bring differentiation to data management that's not being seen with on-site cloud data management. For instance, Snowflake's cloud data solutions are built atop infrastructure storage from Amazon, Microsoft, and Google Cloud (a subsidiary of Alphabet). Whereas customers that utilize one of these platforms, such as Microsoft's Azure or Amazon's S3, are locked into these storage services, Snowflake allows for the ease of sharing cloud data within a company or with other Snowflake users. In other words, business data is no longer confined to a single storage solution with Snowflake.

The company is also a big-time disruptor on the payments front. Whereas most SaaS companies are built on the subscription-based model since it guarantees a certain level of cash flow and discourages customer churn, Snowflake operates as something of a pay-as-you-go model. Users pay for the data they store and the number of Snowflake Compute Credits they use to compute. This should allow businesses to have more control over their expensing, which Snowflake's accelerated sales growth suggests its customers obviously like. 

A person holding a magnifying glass over a company's balance sheet.

Image source: Getty Images.

Berkshire's SaaS cloud IPO bet likely has nothing to do with Buffett

With Berkshire Hathaway sitting on a record $147 billion in cash entering the third quarter, Buffett and his team obviously had some excess cash to toss around. But spending up to $573 million on the hottest cloud IPO of the year almost certainly had nothing to do with Warren Buffett.

One of Buffett's greatest attributes, beyond just being patient and allowing his investment thesis to play out, is that he has a relatively narrow investment scope. This is to say that he focuses most of his attention on financials (e.g., bank stocks) and consumer staples. Digging into tech stocks has never been Buffett's forte, nor would he ever invest in something that he couldn't understand or explain in one sentence. The Oracle of Omaha has even said that Berkshire's investment in Apple, currently the only tech stock in the company's investment portfolio, was made due to Apple's sustainable branding advantage and not because it's a tech company.

Berkshire Hathaway's investment in Snowflake has all the hallmarks of Buffett's investment lieutenants -- Ted Weschler and Todd Combs -- being responsible for pulling the trigger. This is noteworthy because we're seeing Combs and Weschler become more involved in Berkshire's day-to-day investments. Buffett has stated that he had nothing directly to do with Berkshire's stakes in Kroger, Amazon, and Teva Pharmaceutical Industries, and given how critical the Oracle of Omaha has been about gold over the years, it's highly unlikely he was the driving force behind the second-quarter stake taken in Barrick Gold.

Put simply, Snowflake's business model wouldn't be easy for an old-school investor like Buffett to understand. The reason he usually avoids innovative tech companies is because he's used to analyzing companies with hard or tangible assets. An asset-light business model like Snowflake, which is built on scalability, isn't something Buffett would directly buy into.

Yes, Berkshire Hathaway owns a stake in the hottest cloud IPO of 2020. But no, Buffett is almost certainly not behind this investment.