Transforming data teams into growth drivers: Standard implementation playbooks for dbt and Hightouch rollouts by Red Ventures from Coalesce 2023

"We're really passionate that the data warehouse should be the source of truth across the business."

- Tejas Manohar, Co-Founder and Co-CEO of Hightouch

Tejas Manohar, Co-Founder and Co-CEO of Hightouch, and Brandon Beidel, Director of Data Engineering at Red Ventures, explain how they utilize the modern data stack across various client properties and portfolios. The conversation centers on how the company creates a playbook using dbt and Hightouch and how this playbook is applied across various brands under the Red Ventures portfolio.

Red Ventures uses the modern data stack to optimize marketing strategies

Red Ventures, a company that manages multiple brands, leverages a modern data stack to optimize its operations. Using tools like dbt and Hightouch, they have built a playbook that allows for the streamlined application of data across all brands under their management.

Brandon explains, "Our data team's responsibility is to model data from customer experiences: their website, their cart flows, their marketing experiences...and aggregate them into various data warehouses. We model that with dbt." He also highlights how allowing marketers’ access to this data enables them to move faster and integrate with multiple media channels without needing permission from the data team.

Tejas discusses how Red Ventures is going "beyond the typical use case of data warehouses" by activating data for marketing use cases, made possible through Hightouch's reverse ETL capability.

Challenges and solutions in implementing a package-based playbook

Implementing a package-based playbook across multiple client environments presents unique challenges. From dealing with different SQL dialects to varying naming conventions and locations, the Red Ventures team had to develop practical solutions to make the transition smoother.

Brandon mentions some of the strategies they adopted: "If you think there's a chance you're going to need to apply logic across multiple warehouses, use cross-[database] macros... What are some basic assertions that you want to make? What are the primary keys? What can be null? What can't?"

By separating concerns and creating reusable components, the data team was able to produce reusable components that marketers could leverage. As a result, they could scale across multiple channels and clients more efficiently.

Streaming capabilities in Reverse ETL 2.0

Hightouch introduces a new technology called Reverse ETL 2.0, which offers streaming capabilities in Reverse ETL. This new feature allows you to save a series of SQL queries directly into Snowflake, enabling incremental processing as new data comes in.

Tejas Manohar explains: "We've actually unlocked people to be able to do Reverse ETL off [the] data warehouse, with latencies of as low as 15 seconds, or a few seconds, for inserting into raw tables…”

This new capability allows companies to unlock streaming and lower latency workloads off the data warehouse, reinforcing Hightouch's belief that the data warehouse should be the source of truth across the business.

Tejas and Brandon’s key insights

• Red Ventures has created a playbook using dbt and Hightouch to manage their various client properties and portfolio brands
• The company uses the data warehouse beyond typical use cases, powering marketing use cases, personalization, optimizing digital advertising, etc.
• The use of dbt's features allows Red Ventures to create a playbook that can be rolled out across all the different brands in its portfolio
• Hightouch has released a new technology called Reverse ETL 2.0 which offers streaming capabilities in Reverse ETL
• Red Ventures uses various methodologies for media modeling, including match market testing and time series detection