Now that you’re prepared to build your modern data platform with Snowflake, you need to determine the best path to get there. Here are two recommended methods to implement a Snowflake project that will get you to the same end state:
- End-to-End Path: Build your end-to-end solution, from raw layer to data hub to data warehouse.
- Analytics-First Path: Build your Snowflake data warehouse and analytics environment first, and work your way back in future phases.
This blog post originally appeared as a section of our eBook, “Snowflake Deployment Best Practices: A CTO’s Guide to a Modern Data Platform.” Click here to download the full eBook.
End-to-End Path
The end-to-end path will be a higher initial investment, but you will enable both integration and analytics by the time you are finished. When you know what you want and have a fairly stable business environment, this path will ensure you avoid accruing technical debt by continuously putting off work that needs to be done.
Identify a key subject area, such as financial reporting, and bring in only that data throughout the data flow. By the time you finish your first phase, you will have enabled financial analytics, integration of financial data into other applications, and a full end-to-end solution that can be expanded upon with more data in future iterative phases.
Analytics-First Path
The analytics-first path takes a different approach and is likely the most common one you will see in Snowflake POCs. This path is a lower initial cost from a services and work standpoint, and it can produce business value in a much shorter time frame. If you need to show results to get buy-in to a larger project, the analytics-first approach will allow you to load raw data from your source applications and leverage Snowflake views for analytics tools to consume.
The trade-off with this approach comes in the form of an incomplete solution. By identifying several key subject areas, such as both financial and sales reporting, you will be able to get buy-in from multiple stakeholders in the organization. The temptation will be high to continue to add more functionality and more data as business demand grows. Working your way back to create a full end-to-end solution, however, is the right next step. The implementation costs will be higher, but you will be able to create a better-governed data platform, reduce Snowflake costs, and allow for application integration by following the plan.
Both options will lead you to success. If you have buy-in from business leaders and feel comfortable with your vision, building the end-to-end path will allow you to create a strong foundation at the beginning. If you need to show fast results or need to pivot quickly, the analytics-first path will help you sell the value of the solution by showing stakeholders results in dashboards before making a bigger investment up front. Both require a plan to see the entire vision executed but take different routes to get to the end state.
Conclusion
Knowing where to go in your journey is worth the investment.
Snowflake implementation partners, such as 2nd Watch, know how to leverage Snowflake for its strengths and other systems to complement and augment. We know the ecosystems, such as Azure and AWS, and we know how to help you put a strategy in place so Snowflake doesn’t become an analytics silo. We will help you select the right path that creates success for both you and your organization.
If you’re in the process of implementing Snowflake, wondering how it might fit in your organization, or just dreading your next daily ETL load, we would be happy to schedule a free whiteboarding session to brainstorm some ideas together. Click here to get started.
