Transforming the supply chain with Fivetran and Snowflake.
As they were growing rapidly, Redwood Logistics needed a modern data stack that could support their M&A strategy and provide leadership with an up-to-date, accurate overview of business performance.
2nd Watch helped build Redwood a modern data architecture, using Fivetran’s log-based change data capture to replicate on-premises SQL data into the Snowflake Data Cloud.
More data can be processed and accessed faster with minimal source system impact, while freeing developer resources from fixing broken pipelines. Plus, leadership can better track business metrics in near real time.
About the Business
Redwood Logistics, one of the fastest growing logistics providers in the US, has provided solutions for moving and managing freight for more than 20 years, connecting customers to the power of supply chain management, technology, and the industry’s brightest minds.
The Business Challenges
Building an agile supply chain is nearly impossible when you’re trying to control and manage a million things. Systems, shipping partners, applications, processes, all disconnected, all operating independently, all providing no visibility.
That’s where third-party logistics and transportation management firm Redwood Logistics comes into play. The Chicago-based company partners with its clients to develop custom solutions for every area of their supply chains.
From warehousing to its brokerage and fully managed transportation products, Redwood Logistics’ wide array of solutions required a complex reporting structure that relied on multiple siloed warehouses. The rapidly growing business needed a modern data stack that could support the business’ mergers and acquisitions strategy, providing leadership an accurate overview of business performance in near real time in Tableau.
The Solution & the Business Benefits
Generating 500,000 data points per hour
Redwood turned to a team from 2nd Watch — the global cloud services company that provides business outcome-based consulting services for modern data, analytics, and application solutions — and they helped build Redwood a modern data architecture, deploying Fivetran’s high-volume data replication solution to automate the data replication from its on-premise SQL databases to the Snowflake Data Cloud.
“It took us two weeks to build five connectors right off the bat. This would have taken us three months to build had we not used Fivetran,” says Cole Kardys, VP of data and analytics for Redwood. “Thanks to Fivetran, we were able to integrate our Zendesk service desk ticket reporting system and have it fully functional almost immediately.”
From simple setup to modern data stack
Part of the power of Fivetran’s offering is its ease of setup.
“There’s no manual development needed to create data dashboards to show transaction volumes or current health statistics,” says Matt Norris, database business intelligence architect at Redwood. “Once you get Fivetran set up, you very rarely have to interact with it. As a data loading tool, that’s exactly what you’re looking for.The last thing you want is getting calls in the middle of the night, saying that your data isn’t right. We can now focus on the design and architecture of downstream processes.”
Fivetran now replicates numerous data types in Redwood’s Snowflake Data Cloud, including
- Operational data
- Invoicing and payment data
- TMS data
- Rating info
- Extensive audit data
The data that Fivetran replicates makes it easy for leadership teams to conduct rapid and comprehensive reporting in Tableau, such as showing last year and current year revenue numbers.
Redwood was initially cautious about using Fivetran’s high-volume data replication because the team needed to understand how it interacted with their existing databases.
When the 2nd Watch team joined the effort to re-architect Redwood’s stack, it recognized that a big part of the problem was the vast amounts of data involved with Redwood’s core businesses. The old, home-grown system could only load data maybe once a day. Worse still, the system generated numerous daily failures and became a massive maintenance burden.
“Our proof of concept was crucial,” says Fred Bliss, CTO of Data Insights at 2nd Watch. “We had to show that Fivetran’s high-volume data replication didn’t cause any harm to Redwood’s critical systems that were the core of their entire business. What we were expecting Fivetran to do, it did, which is always nice when you’re implementing a tool for a client. It was a good moment on the project for sure.”
However, once Redwood cleared that hurdle, it was smooth sailing, and now the enterprise application has replaced numerous spreadsheets and ad hoc inquiries with a more modern data stack and data visualizations from Tableau — commonplace in daily status meetings.
“You could produce a ton of custom code to do the same thing Fivetran’s high-volume data replication does, but it’s not going to be as maintainable or as easy to write, and it also will run a lot slower,” says Bliss.
“Fivetran’s high-volume data replication would just pipe that data into Snowflake where we would model it and then Tableau would sit right on top of that model data to expose insights,” says Evi Hatzopoulos, enterprise account manager at 2nd Watch.
“Using tools like Tableau just shows the end picture of why all that work was necessary,” Hatzopoulos adds. “Such as how we move data from all these sources, how we combine it, how we have to clean it up and model it to get to this really clear, concise picture of what’s going on in their organization.”
Focusing on what’s important
Thanks to Fivetran and Snowflake, Redwood can now quickly supply custom and feature-rich data dashboards to internal customers and leadership so they can better track their business metrics. Redwood can access and process more data in a timely manner than before.
Fivetran helped Redwood enable its customers to quickly recognize and adjust to changing market trends. Most importantly, Fivetran helped free up Redwood’s team to spend more of their time on value-add projects.
“Development team resources shifted,” says Kardys, “from making sure our code was up and running towards enhancing our tools and adding features.”