My Thoughts On Snowflake's 3Q23 Results

Another Quarter in the Books

It should come as no surprise that I'm a big fan of Snowflake . Anyone who has heard me talk about Snowflake knows I'm all-in when it comes to evangelizing their mission.

The following are my thoughts around Cleveland Research Company 's latest publication about Snowflake's 3Q23 results and their interviews with Snowflake leadership. These thoughts are mine and mine alone, so take from them what you will, and go grab a copy of their report if you want to read more in-depth.

Growth slowed a bit for SMB, but was offset by Global 2000; moderation expected for 2024

SMB (small and mid-market business) growth slowed by 10% in Q3, but with 28 Global 2000 adds (most of any quarter in 5 quarters), 3Q consumption finished ahead of targets. Like most groups, Snowflake is anticipating slightly lower growth for FY24 (47% Y/Y growth for FY24 vs ~68% Y/Y in FY23).

Financial services leads the way as usual

Financial services grew faster than overall company revenues...again. I'm led to believe that the initial use cases in Financial services are more apparent than for groups in, say, Healthcare/Life Sciences or Manufacturing, and generally come with a slightly larger team of people to carry the torch.

SMB is missing the ball with data sharing

CEO Frank Slootman reported that 22% of customers have at least one stable edge of data-sharing capabilities, up from 17% one year ago. Additionally, 66% of organizations with >$1M in annual usage on Snowflake have at least one stable edge deployed for data sharing. To me, this signifies the continued lack of customer maturity across the board, and particularly across SMB, as customers are still working to get their first team of consumers onboarded and operationally sufficient before getting into more advanced capabilities. I'd expect to see this grow quite a bit more than 5 points each year for the next 2 years.

Data Science and Data Application development are full-throttle

Snowpark for Python has received a ton of fanfare, and Snowflake leadership is obviously very passionate about continuing to make Snowflake the nucleus of all data workloads moving forward. The anticipated promotion of Unistore into Public Preview means Snowpark will be used across transactional and analytical data without requiring any movement, and Streamlit will enable native application developement.

Speaking of Streamlit: Snowflake is reporting a 50% community growth since acquisition, bringing the Streamlit developer community to over 70,000 developers. Impressive.

Manufacturing is still missing

As some of you might know, the Manufacturing vertical within Snowflake used to be tied into CPG and Retail, but was spun out earlier this year into its own vertical. Obviously I'm fairly biased, considering my group treats Manufacturing as a primary customer vertical, but I'm very bullish on the Manufacturing leadership team Snowflake has put in place. Connected Factory and IoT workloads notwithstanding, Manufacturing has a huge need to displace outdated legacy software, and features like Unistore will allow groups to move away from old Oracle hybrid table systems for the first time. Not soon enough!

Roadblocks

There are several challenges I see preventing even greater growth on the platform, which I'll list in no particular order:

Data Operations

Getting stood up with an initial Snowflake deployment to explore features and functionality is easy. This is what locks in initial contracts, and allows Snowflake to grow its net-new customer base. HOWEVER, if the bet is to rely at all on SMB for consumption growth over the next several years, there are a few issues to contend with. Firstly, SMB do not have a team of data personnel to run the day-to-day environment requirements within the business. Internal personnel end up spending the majority of their time in the care and feeding of their environment, as opposed to actually building value. This slows consumption and makes contract renewal growth more difficult to obtain. So whether we're talking about monitoring/observability, data source connections, or environment optimization, it's leaving less time for internal team members to drive ROI and prove investment theses.

Vendor Ecosystem

The ISV ecosystem for Snowflake is expanding at an unprecedented rate. While this is beneficial for solutions providers, it's leaving customers assembling a motley crew of technologies that go underutilized. Customers have a hard time filtering between ISV technologies that are actually useful, versus technologies that demo well.

Conclusion

Snowflake continues to outpace its competition, and is growing a vibrant ecosystem for innovation. The responsibility will continue to shift toward partners (systems integrators, managed services providers, etc.) in framing outcomes-based narratives as opposed to solely talking about features and functionality. North Labs recently launched Data Concierge in order to help solve for this gap, and we've seen a tremendous wave of outreach for the service thus far.