As I’m sure you’ve heard, we got the year’s hottest IPO today, with Snowflake (SNOW) launching trading at last.
The IPO had originally been set to go off at $70/share. That figure increased several times in recent days, and the IPO ended up launching at $120/share – a huge jump. Even that ended up looking conservative though, as shares soon topped $300 once trading began. Shares settled their first day at $250, up more than 110% from the already increased IPO price.
Incredibly, this leaves Snowflake at a valuation of $70 billion at this time. That’s up more than 5x from when the company raised money earlier this year. And Snowflake’s valuation is up 20x from last year, when the company switched CEOs. It’s been a truly amazing run.
Why are investors so delighted by Snowflake?
Why This Business Is Different
I won’t quote Snowflake’s prospectus extensively as it uses a ton of jargon and buzz words that make it hard to get a handle on the exact nature of the business. However, this extract should be enough to give you a general sense of the objective here:
Our platform solves the decades-old problem of data silos and data governance. Leveraging the elasticity and performance of the public cloud, our platform enables customers to unify and query data to support a wide variety of use cases. It also provides frictionless and governed data access so users can securely share data inside and outside of their organizations, generally without copying or moving the underlying data.
As a result, customers can blend existing data with new data for broader context, augment data science efforts, or create new monetization streams. Delivered as a service, our platform requires near-zero maintenance, enabling customers to focus on deriving value from their data rather than managing infrastructure.
The broad idea here is to give users one cloud where they can store all their data seamlessly, and then manage who has access to things. This solves many problems around having multiple databases, multiple user authentication schemes, and various security and compliance problems that come with a cobbled-together data network.
By contrast, with Snowflake, administrators set up just one unified cloud data network, and can run all their data analysis, processing, science, sharing, etc. from one place. Snowflake also has strong network effects because it allows customers to share data seamlessly with other Snowflake clients.
In the IPO Roadshow, CEO Frank Slootman (much more on him below) lays out the clear difference between Snowflake and most of its rivals.
The difference is that most SaaS companies built their software to be platform agnostic; i.e. it runs the same on-premise or in the cloud. Obviously, this is not optimal for performance as opposed to a native system. Snowflake, by contrast was built cloud-only and the optimization allows it to handle far more data cleanly and quickly that kludged together offerings from rivals. Snowflake’s founders, in fact, deliberately decided that they would never run on-premise, it was cloud or nothing.
Snowflake claims that it is not possible to achieve all five pillars with an on-premise rooted database alternative.
In any case, that “burn the ships” bet worked. As of the end of July, the company had 3,117 different customers. That is more than double the figure it had in July of 2019.
More importantly, however, Snowflake is landing huge clients. According to the prospectus, it already has seven of the top 10 Fortune 500 companies as customers, and 146 out of the whole 500 list. The company is also scoring an absurdly good 158% net retention rate of revenues. This means that customers not only aren’t leaving the platform, instead, they are spending 58% more on services than they did in the previous year.
When you attract many of the world’s biggest companies to your cloud, and then get them to increase their spending at an exponential growth rate, you’ve got a monster on your hands. Forget about snowflakes, this looks more like a avalanche rolling down the mountain.
Snowflake’s Superstar CEO
CEO Frank Slootman is just the right person to lead the company. You may not know Slootman’s name, but he’s one of the more successful leaders in Silicon Valley. He first took over the reins at Data Domain when it had just 20 employees and no revenues in 2001. By 2008, it had gone public, generated a billion dollars a year in sales, and eventually sold itself to EMC at a fine price.
Data Domain wasn’t a fluke either. Once Slootman had time to recharge his batteries, he took over as the new CEO of ServiceNow (NOW). You may have heard of that one – it’s one of the most successful SaaS companies out there. Slootman led ServiceNow through its IPO and a whole lot more – he stepped aside in 2017. Here’s how that played out during his run:
Slootman stepped aside in 2017, with the stock surging fourfold during his run. And he laid the foundation for further success — NOW stock is now up to $400 as of this writing. Anyone that bought the IPO and held on regardless of valuation and the occasional soft quarter had a home run stock on their hands.
Fast forward to 2019 and Slootman was once again restless and ready to take over a big mission. However, as he said, following ServiceNow, the bar was high for anything that could top that. Luckily, Snowflake’s Board of Directors was not pleased with the company’s strategic direction. Despite being reasonably successful at the time, Snowflake pushed out its then-CEO and had Slootman step into the role last year. Why’d Slootman take the post? As he said:
Snowflake is a special company, they just don’t come along often. I could not resist having an opportunity to bring something to market this promising.
And why did Snowflake boot out their old CEO even though the company was reasonably successful? Slootman says it’s because Snowflake wasn’t focused enough on pure execution in making its product great and driving adoption. From the same interview I cited above, he said that:
“We [Snowflake] were undoubtedly slow to shift gears to pure scale execution, and we are dealing with that now every day and we will from here on out.”
This is, it’s worth noting, exactly the same formula that Slootman used at Data Domain and Servicenow. About Data Domain, Slootman wrote that:
It is remarkable how little our strategy changed from dollar zero to a billion in sales. The most important thing we did throughout the journey: resist the ever-present temptation to muck with the strategy.
In his book, Slootman describes how his only driving focus was on building a better and better data storage product that customers would want and would willingly ditch their old tape backups for. He says that most startups fail because they deliver a product that is cool or innovative, rather than one that is geared to the specific needs of actual customers. Specifically, Slootman wrote that:
Many technologies are conceived without a clear, precise notion of the intended use. There is plenty of hoping and praying going on that some new technology will magically find a suitable problem to solve. Often, we think we know, vaguely, in the abstract—but the truth is we have no clarity on how our technology stacks up in that use-case, relative to alternatives.
This fundamental disconnect is often dismissively listed as a “go to market” problem: something to be figured out later. Well, figure it out now—if you haven’t, you are relying on luck. Lousy bet.
Data Domain was successful precisely the mission at all times was building a specific product that customers would prefer to tape. Slootman described spending countless hours with customers watching their data storage systems in action so they could know exactly how to design the Data Domain product to fit that use case.
Same with ServiceNow’s growth from $75 million to $1.5 billion under Slootman’s care. And now, as he recently said, 100% of the focus is on execution at Snowflake, rather than getting sidetracked with non-essential aims.
It’s particularly interesting seeing Slootman’s backstory as it pertains to Snowflake. At Data Domain, the one and only driving goal was to put the tape backup companies out of business. He built a superior data storage business and got paid accordingly. Now, arguably, Snowflake is doing it again, building the premier cloud data storage and processing infrastructure platform.
Snowflake is particularly focused on building its sales network now, which is why the company is reporting significant operating losses. But when you’ve got Slootman running your sales network, do you want him to hold back any spending here? As he wrote about Data Domain:
Our version of victory was a great salesperson quitting the competition and joining our band of brothers. Breaking their will to fight (prompting “surrender”) was one thing, but getting them to defect outright and rally to our cause—this was crushing for incumbent morale.
There is plenty of resistance in the venture world against building out a direct sales and marketing function because it is so costly. In this regard, we had one super-important thing going for us: high margins. Our product margins were north of 80%.
Notably, Snowflake’s gross margin now is “just” 62%, though that’s up from 53% last year. While that 62% is lower than other SaaS businesses, it’s way ahead of data storage peers that are viewed as more commodity businesses.
Moving to growth, Snowflake is running huge losses. That could be bad if management was not spending its capital wisely.
Remember Snowflake was at a relatively low valuation last year in part because the old CEO wasn’t focused on execution. By hiring Slootman, however, Snowflake brought in a CEO with this vision of the world:
In sales, it was a productivity calculation: we needed to stay ahead of the ratio of sales-expense dollars as a percent of revenues. A motivated, well-compensated sales team can often close business, but if their commission incentives and pricing flexibility are allowed to surrender too much margin, the economics won’t work. Sales has to pay for itself and for the cost of goods, with some margin left over to pay other expenses of the business.
—
We hated raising capital—it was distracting to the business and diluted equity. As time went on, we eased up on this wild animal mentality but we never totally lost it. When EMC acquired us, they were surprised—if not slightly puzzled—to see how tight, disciplined, and cost-conscious Data Domain was. They were perhaps expecting the stereotypical, free-spending West Coast operation without adult supervision. What they got was a lean, mean, fighting machine.
When a jockey has proven himself as tremendously as Slootman has, and he comes out of retirement for one final crowning race, you should seriously consider going along for the ride.
Greatest SaaS Biz On The Market
When I say this is the most important and interesting tech IPO in many years, consider the following:
Snowflake is the red line, and it’s grown faster than any of 39 other leading SaaS companies. In fact, Snowflake has accomplished more in the past year than the average Saas company does in its first ten quarters once hitting a respectable level of revenues. The closest competitor to Snowflake’s speed was Zoom (ZM) (blue line), but even there, Snowflake is clearly ahead – at least up until Zoom got its pandemic bump this year.
I originally was going to have a much longer section here, when I intended to buy SNOW stock post IPO. Alas, the valuation has gotten so crazy that it’s not really worth laying out more of the bull thesis. At least not yet; we can revisit if the price drops. Which brings us to the key context for Part 2…