Building engaging data products: Principles and real-world examples for increased user adoption from Coalesce 2023

"Sigma is the exploratory layer of the modern data stack, and it provides a business intelligence interface that users already know because it is based around a spreadsheet."

- Nate Meiner, Director of Partner Engineering at Sigma Computing

Team members at Sigma Computing, Nate Meinzer, Director of Partner Engineering, and Jake Hannan, Senior Manager, discuss building massive multiplayer data products—with a focus on bringing business teams and data teams together. They also describe how Sigma, an exploratory layer of the modern data stack, can be used to streamline data processes and improve collaboration between teams.

Boosting collaboration between data and business teams through the modern data stack

Nate and Jake highlight the need for better collaboration between data teams and business teams, particularly in the context of managing data within a business intelligence interface. They compare the traditional approach to data management to a "hostile environment,” with a constant back and forth between teams.

"There's this back and forth where somebody has a bunch of data and the business has everything living in Excel," says Jake. Nate explains how Sigma has modernized this process by allowing the business to work directly in Sigma and write information directly back to Snowflake, eliminating the need for Google Sheets or Excel.

By keeping everything within Snowflake and Sigma, the data team can better manage and observe all aspects of the data, leading to more actionable insights. According to Nate, this helped Sigma create a stack that allows people to do everything they need for their job within their platform, leading to greater governance, scalability, and security.

Empowering business users to take ownership of static values

Nate and Jake also emphasize the importance of eliminating the "spectator mode" and giving business users the ability to own and manage static values. They showcased Sigma's input tables, a feature that enables users to write back to Snowflake. Nate argued that this feature puts the users who understand the business requirements in control of the data.

"Today's working environment...the people who know best are the actual business users," says Jake. Nate demonstrates how Sigma's input tables allow business users like himself to update and sync data directly to operational systems such as HubSpot. He argues this reduces the data team's backlog and brings business users into the process of managing their data.

Automating workflows to reduce daily quests and end data exports

Nate and Jake stress the importance of automating workflows to limit “daily quests,” or repetitive tasks, and end data exports. They demonstrate how high-touch tools can fetch query structures from the Sigma API and execute them against a cloud data warehouse, updating fields in operational systems based on the logic applied in the workbook.

Nate demonstrates this by syncing data from Snowflake directly to HubSpot. "It's one less ticket on the data team's backlog," he says. "It's another way of bringing business users into the process of managing their data."

Nake and Jake also discourage the practice of data exports, arguing that when data leaves your platform, you've failed. Instead, they suggest that creating a stack that satisfies as much of the user's workflow as possible can help keep the data within the platform, enhancing security and reducing friction between teams.

Nate and Jake’s key insights

• Sigma provides a business intelligence interface that users are already familiar with—one that’s based around a spreadsheet and built into their data warehouse
• There is often a disconnect between data teams and business teams, leading to inefficient processes and a hostile working environment
• Sigma can help to modernize data architecture and eliminate ungoverned manual work
• Sigma allows business users to write information directly back to Snowflake, eliminating the need for Google Sheets or Excel
• The platform can also trigger transformation pipelines directly from within Sigma, without the need to leave the interface
• Sigma can be used to automate daily tasks and send relevant updates to users
• It is important to end data exports and keep data within the platform for security and governance