8 Takeaways from Snowflake Summit 2022

If you are in the data business, then you must be aware of the 2022 Snowflake Summit that occurred last week. Snowflake has been at the forefront of technology and data innovation, inspiring entirely new markets for nearly ten years. The summit is not only a venue for people to share their learnings in the data business but also a celebration of all the achievements made by the industry.

While the speakers shared great content and wisdom across more than 300 sessions in total, a few themes stood out. Below are my eight key takeaways about how these industry leaders, product visionaries, and technical specialists use the power of data to redefine their businesses and tackle the data challenges faced in their industry.

1) Snowflake is promoting cross-cloud collaborations

Snowflake envisions a new and seamless way for organizations across industries, public clouds, and the globe to securely share and collaborate on data and analytics. In their keynote: Go Further Together with Data Cloud Collaboration, Jennifer Belissent, Snowflake’s Principal Data Strategist, and Prasanna Krishnan, Snowflake’s Director of Product Management shared how collaboration via The Data Cloud is transforming industries and enabling solutions to address some of the most pressing challenges, such as global sustainability.

The Data Cloud, which consists of thousands of Snowflake customers and partners, enables any organization to reveal previously-unimaginable insights and create revenue streams with new data products and services. Snowflake showcased its new capability of insight sharing without exposing the underlying data. Data engineers, data architects, and data managers can “ditch ETL” and use Snowflake to share and access data across clouds and regions. Which cloud you are on will not matter anymore because Snowflake Data Cloud will be the data layer on top of it, making it more efficient. Eventually, many developer workflows, front-end, and back-end can be built on Snowflake.

2) Snowflake is building a strong ecosystem play

Among the data industry players, Snowflake is known for its innovative and data-driven partner program – the Snowflake Partner Network. Snowflake teams also leverage their own technology, including Snowpark, data sharing, reverse ETL, stable edges, and machine learning to power the Snowflake Partner Network program.

During the three-day event, not only did Snowflake showcase the latest technologies, but they also announced two more investments into ecosystem players through Snowflake Ventures. Domino Data Lab, an MLOps company based in San Francisco, and Matillion, a data movement company based in Manchester, UK, received undisclosed funding from Snowflake, further establishing Snowflake’s presence in the modern data stacks. Just earlier this year, Snowflake Ventures invested in data governance companies, Immuta and Collibra, and machine learning company Dataiku, to name a few. Snowflake is trying to level up its collaboration strategy by using Snowflake together with an ecosystem of connected solutions to tackle data governance, management, and observability.

3) Secure data sharing is the foundation of collaboration

Snowflake’s secure data sharing and private exchange capabilities set the foundation for scalable collaboration. New York City Health and Hospitals shared their experience utilizing Snowflake during the COVID-19 pandemic to build their data infrastructure and enterprise data warehouse and to modernize their data sharing capabilities. Through its advanced Data Cloud technology, Snowflake helped New York City Health and Hospitals develop a multi-tiered data cloud repository that facilitates access to data, introduces frictionless governance, improves the data quality and trust, and limits duplication through its zero-cloning technology.

Through Snowflake, New York City Health and Hospitals allowed their data consumers, scientists, and partners to get targeted and secure access to data easier and faster. They also provided executive leaders, our clinicians, and various local, state, and federal agencies with actionable data in response to the pandemic. In other sessions, Snowflake also showed governance features like row-level security and data masking to ensure that the dashboards follow strict privacy guidelines, highlighting the importance of data sharing for any kind of collaboration.

4) Most data teams are focusing on business value

For many organizations, data is critical for achieving situational awareness, optimizing operations, and developing strategic advantage. But to realize the true value of data initiatives, it is imperative to go beyond thinking about saving compute costs by implementing changes to warehouse configuration.

For example, sales and marketing teams are unleashing their data curiosity and augmenting their sales performance with augmented analytics across trade, channels, markets, and products. And they are identifying drivers that contribute to brand and category performance. In the healthcare and life sciences industry, they are building patient 360-degree views to optimize population health management and product commercialization.; In the retail/CPG industry, data teams are developing new products, improving marketing ROI, and tightening collaboration across inventory management and supply chain partners. Companies need to move from routine business intelligence towards decision intelligence, where data curiosity is rewarded with real business impact and improved business-critical decision making.

5) Enterprises are embracing DataOps and MLOps

Many companies are pursuing a forward-looking data stack on Snowflake and a novel approach using DataOps. Companies use DataOps for orchestration and lifecycle release management and for supporting the set-up and enforcement of enterprise-wide policies. This means that data product teams can be both autonomous and innovative while also adhering to governance and security requirements in data access, residency, and encryption. With DataOps, teams can build and deploy a data product and scale to support hundreds of products. DataOps provides enterprises with the working structures, blueprints, cadences, and collaboration needed to drive continual success in each new domain.

Similarly, machine learning operations (MLOps) provide a framework for streamlining the path of machine learning models from experimentation to production with a foundation of strong data and feature pipelines, along with seamless integration with tools required in machine learning workflows. Snowflake has shown that MLOps can help achieve reproducibility, scale, and governance over the organization’s needs. A well-designed MLOps process increases the productivity of all players in the machine learning life cycle, reducing time to production with results that are reproducible and scalable.

6) Everybody is automating as much as they can

By unifying the core capabilities that analytics and data engineering teams need—including data ingestion, transformation, delivery, orchestration, and observability—into a seamless experience on Snowflake, many teams are able to virtually eliminate the time they spend managing and maintaining data pipelines and devote more time to the differentiated, specialized work that scales the business. Harveer Singh, Chief Data Architect at Western Union, together with Matt Holzapfel, Head of Corporate Strategy at Tamr, shared how they were able to automate customer data mastering and develop 720-degree views of Western Union’s customers. Western Union is leveraging Tamr’s data mastering platform with data stored in Snowflake Data Cloud to accomplish this task, which entails cleaning and enriching millions of customer records.

Security compliance teams often spend too much of their time on manual processes. Automating activities such as evidence gathering, key risk indicator alerting, and compliance heath dashboards can free up time for compliance teams to focus on adding significant strategic value to their organizations. Joshua McKibben and Cameron Tekiyeh from the Snowflake team showed the importance of leveraging the Data Cloud to automate continuous monitoring of essential compliance and audit activities.

7) Better data infrastructure means better security

Data lakes can store unlimited amounts of data. But access to unlimited data is not without its problems. And solving these data problems is the critical step in modernizing security operations. Security Operations Centers (SOCs) can be overwhelmed by the variety and volume of data they have to ingest and analyze, the tools they need to manage, and the incidents they need to investigate. And it’s further complicated by the fact that the data is siloed across security and IT products, enterprise data, and threat intelligence.

The emergence of Compliance Operating Systems has allowed security compliance professionals to finally move towards data-based artifacts to fulfill their compliance needs. Operating systems that employ multiple data sources can be used to create a data-based security compliance program and allow companies to meet their compliance requirements. Snowflake believes that an open ecosystem is the best approach, and they are powering the latest cybersecurity use cases with that approach. Leading security teams are also mobilizing data on Snowflake like never before. Snowflake is welcoming cybersecurity to the Data Cloud with exciting new workloads that remove limitations and drive better security outcomes, mitigating real threats faster and more reliably.

8) Enterprises are building a self-service culture

Companies can no longer afford to wait days or weeks for answers. New imperatives for BI and analytics are those that empower business users to answer their own data questions. To accomplish this goal, companies need to provide their users with self-service data access on a modern data stack, where users can safely consume data for BI and analytics. But they also must provide it midstream as part of the DataOps workflow where data transformation occurs. To do this effectively, data security and governance are paramount, particularly when sensitive data about customers are involved.

Harel Shein, WeWork’s Head of Data Engineering, shared how WeWork overhauled its data systems and culture amidst incredible growth. The WeWork data platform team rolled out internally-deployed platforms and frameworks, including digging deep into user pains, evaluating tech options, and rolling out a new, resilient data stack that made data democratization a reality with enterprise-grade visibility, column-level lineage, and self-service cataloging. They have enabled self-service transformations to a variety of data producers, consumers, and domain owners to help build a truly collaborative data culture.

Conclusion

Overall, the Snowflake Summit did a great job highlighting some of the pressing challenges faced by the industry and bringing forward some potential solutions — both technical and strategic — to address them. It’s obvious that the world’s leading organizations recognize the value of data and are looking to maximize their potential to grow their businesses. Some of these same organizations use Tamr to master their business-critical data to better serve existing customers, expand their business, and help reduce risk.

‍