Benzinga, a financial media and data company, is known for its real-time market news, investment tools, and platforms like Benzinga Pro, supporting technical analysis for traders. The company has scaled rapidly, but it wasn’t until recently that Benzinga began investing in more modern infrastructure to better match its ambitions.
The task fell to Reid Hooper, Benzinga’s Director of Data Science. While his title suggests a focus on modeling and machine learning, Reid’s current role resembles that of an acting CDO, as he’s been caught up with doing a lot of overseeing data strategy. He joined in December, alongside a wave of new leadership. The mandate was clear: turn Benzinga’s fragmented data into a coherent, scalable platform.
“I’m kind of looking around, and I’m like, ‘We have no organization to jump into [data science]. Our data is in no state to do data science in a way that's going to be really impactful or accurate.’ It is the biggest issue.”
The Challenge: Fragmentation & Scaling Pains
Before Reid joined the company, Benzinga had multiple siloed analytics teams, each with their own tooling, processes, and data definitions. “I walked into a big mess for sure,” he recalls. “We have 40 instances of [Google Analytics] and we don’t need that many.” The lack of centralized data infrastructure made answering even basic questions a challenge. Something as simple as “how many active subscribers do we have for our different products?” was a headache to answer, as Benzinga was juggling multiple subscription management systems.
“Our investors are asking, ‘What the heck, how do you not have that number right now for us?’”
Using Segment for event tracking and piping data between systems only made things worse. Segment events flowed into BigQuery, and then needed to be modelled downstream, but they often overlapped or conflicted with data from other sources, creating confusion instead of clarity. This wasn't just a matter of cleaning up a few dashboards - it was an ecosystem problem. Legacy analytics tools, ad hoc instrumentation, and years of one-off implementations had left Benzinga with significant technical debt.
The Solution: SQLMesh & Tobiko Cloud
Reid had a clear vision for where he wanted to take Benzinga. It started with building a modern data platform that could keep up with the company’s day-to-day needs and give every team reliable data tailored to their specific use cases.
“We’ve got several different business lines,” Reid explained. “For example, the two biggest ones are our main website—where you can look up news stories and tickers—and Benzinga Pro, which is geared more toward technical and real-time analysis for active traders.”
The audiences of Benzinga’s different product lines might be different, but the content often overlapped. That’s where the data started getting messy. The team needed a solution that could track engagement across both lines, while still being flexible enough to handle their unique scheduling and analysis needs. Given his prior experience, Reid assumed they’d stick with similar tooling. But his lead data engineer, Cortland, introduced him to SQLMesh and Tobiko Cloud early in their research, and it didn’t take long for the Benzinga team to spot why Tobiko’s products stood out from the rest.
- Column-level verification: Instead of catching errors in production, SQLMesh flagged them at the column level, avoiding costly and time-consuming errors that previously went undetected.
- Environment engagement: SQLMesh made it easy to manage isolated dev, staging, and production environments and clean them up when no longer needed. This gave Reid’s team the confidence to experiment.
- Hot-swapping between dev and prod: SQLMesh allowed for seamless backfills in development and fast promotions to production, which is important for the Benzinga team’s ability to handle large datasets.
- Cost and time savings: SQLMesh’s way of handling dependencies and backfills made a big difference in performance and cut down Benzinga’s costs. “We work with big volumes of data,” Reid noted. “So not having to reprocess everything all the time definitely saved a lot of time and resources.”
- White-glove support: Switching tools can be a challenge, but Tobiko’s fast Slack support and smooth onboarding experience made it feel like less of a leap and more of a handoff.
Implementation & Impact Within a Week
Tobiko Cloud and SQLMesh became production-ready within a week - and the benefits also showed up almost immediately.
“We had our environment and first repo and everything configured probably within a week or less. From there, we’re pretty much going to use this as production because it was already solving the needs we had.”
With column-level awareness, potential breakages were flagged right at the planning stage so that Reid’s team could see how changes might affect downstream pipelines before they happened. For a company processing millions of events per week, SQLMesh’s ability to isolate and backfill in dev environments helped cut a lot of Benzinga’s unnecessary compute costs. There was also the vendor experience - compared to past vendors, the Tobiko team felt more like collaborators.
“There were multiple instances where I posted questions or errors in Slack, and someone from Tobiko’s team would reach out and we’d hop on a call within 10 minutes,” said Reid, “That’s much faster than some of the other partners I’ve worked with.”
What’s Next for Benzinga
The data team at Benzinga is already looking at improving Benzinga’s customer intelligence and creating new revenue-generating data products. They’re also exploring a move to Databricks for heavier data storage and processing needs.
“We definitely do want to implement Databricks, and I could see Tobiko Cloud and SQLMesh playing a role there…we’ll collaborate with the Tobiko team as we go down that journey.”
In the meantime, the team at Benzinga is still pulling together event tracking data from tools like Segment and GA4 into BigQuery, where SQLMesh helps keep everything clean and ready for analysis across different parts of the business. For Reid, the foundation is just the beginning. “We have APIs that we sell to quant firms…I think we have the opportunity to actually create new products to sell. And that is also very exciting for me and the team.”
Conclusion
For Benzinga, adopting SQLMesh and Tobiko Cloud was more than just a tooling upgrade - it was a foundational rebuild. What started as an urgent need to fix data fragmentation evolved into a platform for scale and innovation. From faster backfills and column-level model planning to smarter environment management, Reid and his team can now give leadership the answers they’re looking for, support cross-functional goals, and build products with clarity. Benzinga finally has the tools and support to scale with confidence on a platform that’s equipped to work seamlessly with them.
