Berkshire Hathaway Inc., an American multinational conglomerate holding company led by Warren Buffett, has never participated in an IPO offering for over 50 years. However, one outstanding business called Snowflake broke its tradition.

Snowflake is a cloud-based data warehousing business, that became the largest-ever US software IPO in history. But, how could they make the largest-ever IPO offering? What is Snowflake?

What is Snowflake?

Snowflake is a cloud-based data warehousing company based in California and founded in 2012. It provides a wide range of services to help customers get the most out of their store data. With its versatile, scalable, and effective software-as-a-service solution, organizations may save time and money on their data analysis.

Snowflake offers various services through the platform such as data engineering, data lake, data warehouse, data science, data applications, and data sharing. In short, Snowflake helps companies to get more insights from big data more effectively by making data processing and analysis easier. As you can see in the image below, these services run across Google Cloud, AWS, and Microsoft Azure.

You can find the product description to understand Snowflake here.

The history of Snowflake

Benoit Dageville, Thierry Cruanes, and Marcin Zukowski, three data warehousing specialists, created Snowflake Inc. in San Mateo, California, in July 2012. The name “Snowflake” comes from their love of snow sports.

In 2012 when they started Snowflake, the data warehouses are rigid, expensive, and difficult to use. In addition to that, on-premise was the predominant data storage method. The founders saw a challenge here, so they decided to revamp the industry by creating a cloud-based data warehouse. Their new data warehouse needed to be speedy, easy, cost-effective, and elastic.

The name of Snowflake was not known for the first 2 years since they started. However, after they appointed Bob Muglia, a former Microsoft executive, as CEO of Snowflake, the company’s name became known at once. In spite of only 80 clients in June 2015, they had 3400 customers as of 2020, including industry-leading companies such as Capital One, Adobe, and Door Dash.

The Biggest Software IPO in History

Snowflake achieved the biggest-ever software IPO. But why could they do this?

Large market opportunity for cloud data platform:  Management claims the total addressable market for Cloud Data Platform is $81 billion and the markets for Analytics Data Management and Integration Platforms and Business Intelligence and Analytics Tools is $56 billion by the end of 2020 and $84 billion by the end of 2023.

Data Source: https://blog.publiccomps.com/snowflake-s1-ipo-teardown/

Firstly, the market size is growing rapidly in this industry. The market since of data warehouse-related industry is $56B as of 2020 and is estimated to increase to $84B by the end of 2023. According to the reference above, the market size calculation included not only a data warehouse but also data visualization. But, since Snowflake is trying to enter data visualization service as well so this is the total addressable market that Snowflake targets.

The market trend is not the only tailwind for SnowFlake. As the founders of SnowFlake have pointed out from the outset, the growing volume and complexity of data around the world are making it impossible to handle the traditional on-premise approach, creating a strong demand to move from on-premise to the cloud-based data warehouse. According to IDC, the cloud computing rate will be 30% in 2020, while it is expected to reach 50% in 2025.

The annual revenue growth of Snowflake is also outstanding. The revenue increases by 121% yearly. This sales growth shows that the market trend is favorable and Snowflake has a strong competitive advantage.

Market trends, excellent founders, and sales growth resulted in the biggest-ever software IPO.

Why should you use Snowflake?

The importance of data is increasing

As technology is evolving over the decades, the amount of data that a company can use is increasing sharply, so the word”Big data” is a buzzword in the recent business environment. From the 2010s to the 2020s, the data that companies deal with has become increasingly complex and voluminous, stored in a variety of locations. As a result, there is an accelerating need to visualize and effectively utilize this data for business purposes.

According to IDC, there will be 175 zettabytes of data by 2025, representing a CAGR of 27% from 33 zettabytes of data in 2018. 

Source: https://blog.publiccomps.com/snowflake-s1-ipo-teardown/

＊1 megabyte = 1,000,000
＊1 gigabyte = 1,000,000,000
＊1 zettaabyte = 1,000,000,000,000,000,000,000

According to IDC, the data handled by companies in 2018 was 33 ZB and is expected to reach 175 ZB by 2025, with an average annual growth rate of 27% through 2025. Zettabytes are not at all what we can imagine, but it is easy to imagine that it is many levels above the MB and GB units we normally use in daily life.

What does Snowflake offer?

Snowflake offers several services through its platform to leverage the data that your company has. But, what can we do and what is necessary to use the data effectively? Let’s see what Snowflake offers to understand it!

Snowflake mainly offers 5 services that are put over the blue box in the image above: Collaboration, Data Engineering, Cybersecurity, Data Science & ML, and Applications. And 3 basic services underlie those 5 services: Dat warehouse, Data lake, and Unistore. Snowflake’s services are supported by a good infrastructure that is illustrated below those: Cross-cloud snow grid, Intelligent infrastructure, Elastic performance engine, and Optimized storage.

Snowflake services

Collaboration

Organizations can eliminate data silos and the complexity involved in securely accessing and distributing data throughout your business ecosystem with Snowflake. Any business unit, partners, and consumers can manage or access the data stored in the cloud even if they are not on Snowflake. Furthermore, you can reduce the risks and costs that may occur if you use the traditional data-sharing method with Snowflake. Of course, even if you share your data globally, it is secured and well-governed so you do not have to worry about data security.

Since you can manage the right to access your data, Snowflake allows you to monetize your data on the Snowflake marketplace. You can purchase other parties’ data to analyze or sell your data on the platform to make money. You can easily reach potential customers via the platform.

Data Engineering

Data engineering means the building of systems to enable the collection and usage of data. 

Data engineers can eliminate capacity planning and concurrency management, spend little to no time managing infrastructure, and instead concentrate on creating dependable, enterprise-ready data pipelines.

Structured, semi-structured, and unstructured data intake are all handled by Snowflake in batches and continuously. Data engineers can use the data more efficiently and smoothly thanks to this. Moreover, because of its simplified data transformation and architecture, less time in managing infrastructure is required.

No matter what coding preferences your developers have, your developers can collaborate thanks to Snowpark, Snowflake’s developer framework, and then native support for ANSI SQL.

Cybersecurity

Since you will manage big data, the issue of security always comes together. If someone breaks into your data server, it will be a huge problem for the company.

To better secure data management, Snowflake enables us to monitor data more easily. Higher-fidelity alerts and higher credibility threat detections achieve worry-less data management. And you can unify your logs and enterprise data in a single place. Hence, the cost of security can be reduced.

Data Science & Machine Learning

The biggest benefit you can get from big data is insights that can be brought by data science or machine learning.

Any structured(JSON, Avro, ORC, Parquet, or XML) or unstructured data can be transformed and used for machine learning to get insights for your business. Furthermore, the data source can be from a third party that can be extracted from the Snowflake marketplace.

You can also combine the machine learning tool of your choice with data stored in Snowflake. So your data scientist can analyze and get ML-powered insights effortlessly.

Data science use cases include financial forecasting, account propensity scoring, and ML-driven product strategy.

Applications

You can focus on building and using data-intensive applications since Snowflake handles the infrastructural complexity. Snowflake is managing maintenance, compute resources management, administration, networking & encryption so that your teams can work on what actually matters to the business.

Applications can be anything. For example, applications for IoT, machine learning, etc… These data applications can run on the cloud servers such as Google Cloud, AWS, and Azure.

Data warehouse

The goal of a data warehouse is to centralize all data for fast analytics & reporting with less maintenance cost. While traditional data warehouses had issues such as slow performance, Snowflake’s data warehouse enables you to use your data sources effectively.

In short, with Snowflake, you can overview the big data that you own in a short time and analyze it for business insights.

Data lake

The data lake’s role is to provide a single source of truth, storing the raw data in other words. It makes data available for a wide array of users and use cases.

Snowflake’s data lake simplifies the data pipelines, resulting in reduced complexity. Hence, security and reliable data storage are achieved. Surprisingly, this service is offered as a service without maintenance.

Unistore

Snowflake unveils Unistore, a brand-new workload that offers a cutting-edge method for combining transactional and analytical data on a single platform.

In traditional data warehouses, transactional data is stored separately. Therefore, if you want to utilize the transactional data for analytics, you had to copy the all transactional data and gather them in one place. This resulted in redundant data sets, delayed access, and danger to security. However, with Snowflake, transactional data and data for analysis are stored in one place. There are fewer systems to manage, immediate data access, fewer data movements, and stronger security.

Snowflake Infrastructure

All of the services described above are built on excellent infrastructure. Snowflake’s infrastructure removes all concerns related to data administration. The layers of Snowflakes’ platform consist of Snowgrid, intelligent infrastructure, an elastic performance engine, and optimized storage. These are quite technical, so I won’t go into details here, but these technologies allow for excellent data management and data utilization.

What are the advantages of Snowflake?

1. Cloud-based SaaS model requires no infrastructure and is easy to deploy

Snowflake is a cloud-based SaaS data warehouse that requires no infrastructure and no time or effort to set up.

With a 30-day free trial period, Snowflake is one of the easiest services to deploy.

2. Easy to scale up and down while keeping costs low

Snowflake’s virtual warehouses respond by launching the size and number of virtual warehouses as needed, making it easy to scale up and down as needed.

In addition, up to 10 virtual warehouses can be launched simultaneously, allowing for horizontal scaling (multi-clustering).

Another major advantage is that the second-based pay-as-you-go fee structure leaves little room for significant changes in costs associated with such resizing.

3. Non-engineers can perform analysis

In addition to enabling analysis with SQL, Snowflake supports intuitive operation with GUI, so that even non-engineers can participate in the analysis.

Although SQL is indispensable for large-scale analysis, it is a very reliable feature for users who will be working on analysis.

4. No need for data conversion, which reduces management effort

When analyzing data from a service, there are many cases where the data obtained from the API is semi-structured data. In such cases, Snowflake allows the data to be loaded and analyzed without the need for conversion.

Furthermore, by skipping the conversion procedure, data can be easily aggregated in a data lake-like fashion. This is expected to prevent data loss and improve quality from a security perspective as well.

5. Easy to share data

Another major advantage of Snowflake is that role-based access control makes it easy to share data based on proper authority management.

For example, if you want to share data with a customer, there is no need to copy or move the data. As long as you have roles for your customers, you can easily grant the same permissions to multiple users.

6. Reduces errors and backup time

The time travel feature prevents loss that can occur due to human or system error.

Some editions allow up to 90 days of time travel, and with data consolidated in Snowflake, there is little need for manual backups in normal times.

Innovative pricing system

Traditionally, data warehouse services adopted a subscription pricing model which makes the consumers pay for things they don’t use. Snowflake revolutionalize the pricing model for this industry by adapting the utilization pricing model.

The strategy used by Snowflake is to separate storage from computing power into two separate features and offer consumers the freedom to choose how much of each they need. With Snowflake, you may select a certain amount of storage and then change it as needed. The same is true for computing power; if a project you have coming up calls for more than you now have, you can easily upgrade and then decrease the amount after it’s over.

Snowflake Application Examples

Deliveroo

Deliveroo, a London-based food delivery company funded by Amazon, has implemented Snowflake to enable real-time decision-making.

Previously, the company’s ETL and analytical workload increased from the busy weekend to Monday morning, when the company’s management team could not see data in real-time and analysts were unable to plan for the following week.

Snowflake’s native support for semi-structured data and multi-cluster with virtual warehousing solved this challenge. Currently, they are also implementing measures to help improve delivery efficiency using logistics algorithms.

Reference: Deliveroo Delivers with Real-time Data

Domino’s Pizza

Domino’s Pizza Enterprises Limited (DPE), Australia’s largest pizza chain, credits Snowflake for improving data collection and flow.

Originally a data-driven company, DPE has a vast amount of customer data collected from each of its regions around the world, and Snowflake has made it easier to access and analyze that data.

“With the shift to digital, data is more accessible. The resulting data analytics give us a level of insight into customer behaviors that we have never had before.”

Michael Gillespie, Group Chief Digital and Technology Officer, DPE

Reference: Pizza Made Easy at Domino’s: Using Digital Data to Offer Customers Better End-to-End Experiences

Capital One

Capital One, a U.S. company in the credit card, Internet banking, and other financial services business, saw a 4-6x performance improvement with Snowflake.

The company had a complex environment with hundreds of terabytes of data and over 200,000 tables that were being queried by thousands of users simultaneously.

Snowflake responded by scaling out to a multi-cluster virtual warehouse. Another factor in the implementation was that Snowflake met the security and risk requirements required for a financial institution.

Reference: Enabling the Future of Banking with Snowflake

Conclusion

Snowflake is a cloud-based data warehousing business, that became the largest-ever US software IPO. This historical achievement came true thanks to the outstanding executive members and its sophisticated services.

Nowadays, more and more companies are realizing the importance of data utilization for business purposes. If you are interested in data utilization for your company, contact us. We can help you leverage your company’s data to go to a higher level.