On December 3rd, Snowflake, a hot tech company, announced a loss of $1 per share in the last quarter, three times more than what analysts expected. However, Snowflake’s stock price surged by 16%, followed by another 14% increase the next day, making it a company with a market value of over $100 billion.
When Snowflake went public on September 16th, underwriters priced it at $120, which was a 50% increase. However, less than three months later, the stock price is $387.
Investors tear up the target prices given by major investment banks and flatten the shorts, this is exactly the software industry’s version of Tesla.
Is it true that Snowflake is really that strong?
One, Link to heading
Actually, I knew about Snowflake last year while searching for an affordable big data analytics solution.
In a nutshell, Xuehua has transformed expensive data warehouses into efficient cloud services.
My friends always complain about Excel freezing and BI tools running large cross-tabs not performing well. If they decide to build their own HBase-Kafka-Spark-Kylin infrastructure, despite being open-source and free, they would actually end up spending a lot of money renting expensive clouds and hiring costly staff.
Seeing the snow so early is a great thing, but I didn’t expect that investors would also have such foresight.
Perhaps they are simply blindly following Buffett? However, this does not make sense, as the price that Buffett reportedly bought at was only $80, which does not necessarily mean that the current price is reasonable.
Do American investors possess such profound technical insights, or are they simply speculating recklessly with their money?
“I think that irrational madness must also have a rational perspective.”
In today’s world, it is difficult to determine whether a company’s stock price is reasonable or not, although you can judge if it is a good company.
Next, I tried to interpret it.
Two, Link to heading
Experts say that the most important thing about Snowflake is that it reduces customer dependence on cloud infrastructure brands. While this statement makes sense, it is not enough to justify the frenzy around Snowflake, especially considering there are so many open source options available.
Experts say that the three major cloud providers all have similar competing products. AWS has Redshift, Athena, and EMR. GCP has BigQuery, and Azure has Synapse. Therefore, Snowflake faces the risk of being crushed by large companies.
But in fact, the differences between them are extremely large, much larger than the differences between Oracle, Sybase, and DB2 back in the day.
The founders of Snowflake came from Oracle, and they created Snowflake after gaining deep insights into the shortcomings of Oracle and Hadoop. The 8 years and patents during this process is now their protective moat.
The snowflake killer feature is its simplicity, efficiency, and unique secret technique of virtual data warehousing. What’s even better is that it’s also very affordable!
Three, Link to heading
Experts will come out and say, “How could it be cheap? The pricing is not low, you see, storage and calculation are also charged…”
I’ll give an example. Suppose I, a poor private equity fund, want to analyze all the detailed data of the US stock market using big data every day, and automatically compare it with historical data, weather data, and political and economic data to find out whether the stocks will rise or not.
This amount of data is frighteningly large, and according to traditional IT practices, it would require an investment of billions each year. Even major investment institutions don’t have enough funds to handle it.
How can we play with snowflakes?
Because the storage in cloud computing is very cheap, I can store a lot of data. Every day, I can temporarily call ready-made stock data and political, economic and weather data from the Snowflake Data Marketplace.
“Snowflake separates storage and computing. I can rent its computing power temporarily by running the scripts I wrote after the end of each day’s trading, and I don’t have to pay anything at other times.”
If I need to use money to buy time and make it count faster, I can also temporarily rent Snowflake’s self-storage. If I want to do high-frequency trading, it is also possible to analyze real-time during trading hours through pre-machine learning algorithms.
Four, Link to heading
Although this example may be a bit exaggerated, it explains a concept.
This concept is quite similar to Salesforce’s success. Each customer has different reasons for using it, and they all have a need to save money, but they are willing to spend money on something that is sharp and effective: paying only for simplicity and high efficiency, ultimately resulting in lower overall costs.
Everyone knows that Oracle, which once had the most comprehensive On-Premises commercial suite, was overturned by cloud computing and declined. Snowflake seems to have come with a mission to disrupt the previous generation of data analysis platforms.
After all, data has become the most valuable asset and core competency for enterprises. The so-called ability to handle how many concurrent transactional databases will become child’s play, while the value of analytical databases will continue to grow.
So, should we buy stocks of Snowflake? I’m not sure, because for a company that hasn’t earned any money yet, it’s really expensive.
However, as cloud-native and SaaS eventually conquer most enterprises, the technology companies that will follow are endless.