Why Organizations Are (Still) Excited About Snowflake

It wasn’t long ago that Snowflake went public as the biggest software IPO in history. As with any single stock, there have been some relatively big ups and downs for the company. Regardless of how it is performing on Wall Street, our customers are still really excited about Snowflake. They’re excited about how they can focus more on getting value out of their data rather than just managing their data infrastructure. They’re excited about how easy Snowflake is to use. And they’re excited about the future vision of what Snowflake can do for their organization.

What Makes Snowflake a Foundational Platform for Organizations?

A fully managed service, Snowflake is more than just a database—and it is also more than a cloud-based platform where you host a database. Snowflake is best thought of as a data platform that supports a variety of different workloads—including data engineering, data lakes, data warehouse, data science, data applications, and data sharing—making it a foundational platform for all your business data needs. It is designed to offer instant scalability and elasticity… and did we mention that it is easy to use?

For many organizations, it is difficult to effectively scale up data solutions over time as their needs (and their data) grow exponentially. Snowflake eliminates those historical barriers through its single elastic performance engine which allows organizations to easily scale up and down compute independent of storage—either with a click of a button or running of a line of code. Organizations using Snowflake can tackle opportunities that require instant horsepower, yet control costs by instantly releasing the extra resources when they are no longer needed.

This makes it easier to break down data silos and share governed secure data in real time for business intelligence and analytics—both internal and external to their organization. Snowflake has positioned itself as a one-stop-shop to store, manage, analyze, and share data seamlessly—which is something many of our customers are excited about, and are still just beginning to realize the full benefits of.

Snowflake offers capabilities for six critical workloads including data engineering, data lake, data warehouse, data science, data applications, and data sharing. Image: Snowflake

Benefits of Snowflake: How It Can Be Used to Enable Your Most Critical Workloads 

There are many ways to benefit from the power of Snowflake, and even if you’re already a Snowflake customer, you’re probably not taking advantage of all the capabilities the platform offers. Here are examples of the types of workloads you can execute in Snowflake that will benefit your organization.

Data Engineering

Building data pipelines with Snowflake allows for minimal data transfer between multiple tools and storage locations and highlights Snowflake’s scalable compute power. You can select data engineering tools from the thriving ecosystem of partners who integrate closely with Snowflake—such as Fivetran for data ingestion and dbt for data transformation—or create pipelines in Snowflake using features such as Snowpipe, Streams, and Tasks using whatever coding language you want.

Using Snowflake’s native data engineering functionality, companies can build repeatable, scalable processes where transformation logic is completely automated to support an analytics solution that a company uses to offer services to customers. For example, Epiq used Snowflake to automate the data ingestion process for an insights product offered to customers, allowing the company to scale the solution to more clients while saving time and money in the process.

Data Lakes

Use Snowflake to leverage existing data lakes by querying external data sitting in locations such as S3, Google Cloud Storage, Azure Blob Storage, or Delta Lake (preview feature). Or you may choose to host a new data lake directly in Snowflake, perhaps right alongside your data warehouse. Make Snowflake the one-stop-shop for all your enterprise data by consolidating semi-structured and unstructured data alongside structured data—reducing the complexity, data storage, and data transfer costs that come with maintaining a patchwork of data storage technologies.

For example, manufacturing companies can use data lakes in Snowflake to store data generated on the modern manufacturing shop floor in its native format, including IoT sensor and other machine-generated data. Whether this data is streaming, semi-structured, or unstructured—it can be stored in Snowflake’s data lake and used alongside structured data for a holistic view of operations, predictive maintenance, data science applications, and more.

Data Warehouse

Leverage Snowflake’s performance scalability to accommodate large data volumes, frequent data refreshes, or real-time requirements. Concurrency issues such as many users querying the data warehouse at once via a dashboard can be easily solved using Snowflake’s multi-clustering capabilities—automatically scaling out to meet the demand and scaling back in once the peak in demand passes.

Read about how iFit built their data warehouse on Snowflake to combine customer data from many sources systems.

While data warehousing can benefit any department in an organization, we’ve seen marketing teams in particular grow significantly in their capabilities after we built a strong data foundation for them in Snowflake. One of our clients, a top computer hardware manufacturer, consolidated their customer data to provide a customer 360 view, generating key metrics such as churn, customer lifetime value, and creating marketing attribution models for more accurately targeted campaigns.

Data Science

Snowflake’s data science benefits are all about streamlining processes, reducing data movement, and providing scalable performance to meet complex and high-volume query needs. Accomplish data model preparation using your DS framework and language of choice, such as Python. Expand capabilities by using Snowflake’s tight integrations with tools such as Data Robot for AutoML.

Data scientists in any industry can leverage Snowflake to explore large volumes of data, accelerate feature engineering processes, and work with their data science tools of choice to solve complex business questions. For example, a healthcare services organization can develop a machine learning classification model in Python featuring natural language processing (NLP) to understand the factors contributing to patient insurance claim approval rates.

Learn more about our Data Science Accelerator.

Data Applications

Snowflake enhances data application development and speeds development timelines by automatically handling many of the data considerations such as provisioning, availability, tuning, and data protection. Applications which are particularly data intensive will benefit from Snowflake’s scalable compute power, providing the ability to crank up performance to meet user expectations.

Healthcare organizations, for example, have used Snowflake to power patient portal applications, allowing patients to engage with the provider and accomplish everything they need to do—including appointment scheduling, communications, and intake—all in one place.

Data Sharing

Share and collaborate on live data sets with both internal and external parties. Use Snowflake data sharing to remove friction from data ingestion processes. If you are receiving data provided by a trusted partner or third-party data provider, Snowflake data sharing can quickly replace file transfer approaches such as FTP file drops, increasing data freshness and eliminating fragile ingestion processes. This same data sharing functionality can be used to receive external curated datasets from more than 200 data providers showcasing their offerings through Snowflake’s Data Marketplace. Or, become a provider yourself on the Marketplace and create new revenue streams by monetizing your unique data sets.

For example, financial and retail organizations have found Snowflake’s data sharing capability to be a game changer when used to ingest the latest exchange rate data in real time, removing a common bottleneck to more frequent analytics refreshes.

Snowflake has never been a one-trick-pony—its value lies in its ability to execute on a wide variety of data workloads. As the platform continues to evolve and grow, so does the potential for organizations in any line of business that use Snowflake.

Why You Should Be Excited About Snowflake

As we look to the future, Snowflake is not resting on its laurels—the platform continues to rapidly expand its capabilities, most notably around data science, data lake, and data engineering workloads.

• Snowpark for Python will streamline data science workloads by allowing Python applications to be executed directly without having to move data out of Snowflake. Expected to be available in 2022.
• Snowflake Scripting enhances data engineering workloads by allowing SQL stored procedures to be written, stored, and executed in Snowflake directly without requiring other layers or languages. This feature is in Open Preview now.
• Unstructured data support bolsters data lake workloads by allowing users to access, load, govern, and share unstructured files such as text-heavy social media conversations, images, video, audio, and industry specific file types such as genomics data in VCF format.

Keep up with the most important Snowflake updates.

I’ve seen clients benefit significantly by taking advantage of Snowflake’s capabilities for workloads beyond data warehousing. In my experience, the organizations that use Snowflake as a foundational platform to launch into new areas such as data science are the ones gaining the competitive edge in their industries. Snowflake’s ability to enable an organization to accomplish all their present and future data use cases in a single platform makes it a leader in the marketplace and I only see its capabilities improving with time.