- Paytronix uses Fivetran Local Data Processing to bring all its data to Snowflake, including customer data coming in near real time via 12 connectors
- Data is transformed in Coalesce as it comes in, and the Paytronix data team has developed ML models to run in Snowpark without the data leaving Snowflake
- Paytronix clients can create customer experiences and campaigns based on insights generated by near real-time predictive analytics
- Company Size: ~300 Employees
- Industry: SaaS
- Destination: Snowflake
- Cloud: Microsoft Azure
- Sources: MongoDB, SFDC, Stripe, Recurly
Customer engagement platform Paytronix helps its clients—more than 1,800 brands in the restaurant and convenience store industries—leverage their customer data to improve the digital marketing funnel and offer customers a frictionless experience every time they visit the store, whether in person or online. This relies not just on having a lot of data, but having the ability to glean useful insight from it quickly and easily.
As the Director of Data Science at Paytronix, Jesse Marshall leads a team of seven, which includes two data scientists and five data engineers. The team works closely with the larger Strategy and Analytics (S&A) team, which provides clients with insights they need to truly engage with guests. Because of this, his own team’s No. 1 priority is to make data actionable for other departments.
Marshall’s team was challenged with collecting, organizing, and deriving insight from data coming from a multitude of sources, running on multiple databases, and in disparate formats. The company was using a tool for data ingestion that was at times unreliable and missed a lot of transactions, which made it hard to trust the underlying data. In addition, it was using a mix of Scala and PySpark jobs for data transformation—custom code, hand written. It became clear that this tool set was not able to keep up with the growing demands of the business, as the increasing scale was putting pressure on the platform, and a lot of time was dedicated to maintenance and break-fix support. Marshall wanted to get ahead of the game and be ready for increased sales and customer demand.
Additionally, Marshall believed the data science side of the business should be more focused on experimentation, where his team would be able to come up with a data pipeline idea, get things to a proof-of-concept phase right away, and then test it quickly.
We see data in so many different forms, so we have to be flexible with how we can ingest it. When we’re doing an ETL, how do we make it so that it’s not a weeks-long project to get the data into the different departments’ hands?
Jesse Marshall - Director of Data Science, Paytronix
The negative impacts on the business included:
- Team unable to take meaningful action on siloed data coming from many sources
- Patchwork of tools made it difficult to build pipelines quickly and easily
- No ability to fail often and fail fast, which is necessary for data science
Why They Chose Fivetran
Marshall began looking for a solution that would enable his data team to operate in the way he envisioned. He was impressed by Fivetran Local Data Processing’s painless setup and reliability, and decided to replace the team’s legacy ingestion tool and start using Fivetran instead to bring all the data they received from many different sources into Snowflake. In addition, an industry colleague pointed him toward Coalesce as a good fit for what he was trying to achieve, and he was quickly sold on the platform’s ease of use and flexibility.
The team had recently begun using Snowflake Snowpark with the goal of eventually replacing the disparate systems the team used for their data science projects. Snowpark enables data scientists to code in languages other than SQL; they don’t have to take data out of Snowflake to run, for example, Python scripts—they can do it directly where the data lives in Snowflake. Paytronix brings that data in near real time over to Snowflake (via Fivetran Local Data Processing), and then uses Apache Airflow to trigger transformations in Coalesce and then trigger the models to run in Snowpark. With Fivetran providing reliable, worry-free data ingest every 15 minutes, and Coalesce enabling faster, automated data transformations combined with the benefits of Snowpark, Paytronix can now do real-time predictive modeling, and at scale.
The company is able to offer its clients real-time information about their customers’ activity. This allows someone on the marketing team of a Paytronix client, such as Peet’s coffee, to pull the most up-to-date information about their customers when running a campaign, or offer tailored 1:1 messaging that creates a strong personal connection with the brand and results in higher levels of engagement.
Building a foundation for the future
But it wasn’t just Marshall’s smaller team that was able to directly benefit from adopting Fivetran and Coalesce. The analysts on the larger S&A team, frustrated by the time it originally took the overworked engineering team to build pipelines, had gotten in the habit of using the company’s BI tool, which enabled them to create persistent derived tables, as their own ETL tool. But over the years this had led to a lot of confusion around metrics because there were many sources of truth. With Fivetran, Coalesce, and Snowflake working together in one seamless, best-of-breed technology stack, Marshall’s team has shortened the length of development time so that analysts are no longer creating ETLs, and Paytronix has gone back to one source of truth. And because the analysts can understand the code and dive deeply into it to see the data lineage, they have more trust in the data.
For Marshall, all this progress has helped get his team to a place where the important work can now truly begin, and they can offer real business value to the larger company: building new IP, new features, and new predictive models to help Paytronix’s clients offer the best possible experience to each and every one of their customers.
Coalesce and Fivetran enabling faster, automated data transformations coupled with the benefits of Snowpark means that today Paytronix can do real-time predictive modeling, and at scale.
- Faster and smoother time to insights with Fivetran for ingestion and Coalesce for data transformation
- With Coalesce, two new team members able to complete high-profile transformation in one month, whereas before the entire team spent 6 months without much progress
- Data science/AI initiatives executed with near real-time data and the ability to iterate frequently and quickly with Snowflake Snowpark