DonorsChoose Makes the Right Choice With a Modern Data Stack
To leave behind a system of Excel spreadsheets and SQL, DonorsChoose brought on Fivetran, Redshift and Looker. Since the initial implementation, the business has connected to Salesforce, Zendesk and Heroku and migrated its production database to AWS RDS. The organization’s daily dashboard displays total revenue, number of acquired users and year-over-year changes, which helps daily in making strategic business decisions.
- Data Pipeline: Fivetran
- Connectors: Heroku, PostgreSQL, Amazon S3, Salesforce, Zendesk
- Destination: Amazon Redshift
- BI Tool: Looker
DonorsChoose is a US non-profit that makes it easy for anyone to help a classroom in need. Its mission is to help students in every community have the tools and experiences they need for a great education. Teachers post classroom projects for supplies or experiences that donors and partners can contribute to. Mohammad Radiyat is the sole Data Scientist at DonorsChoose, handling the data pipelining and engineering and machine learning and Andi Muskaj is the sole Business Analyst. With a small data team, the organization needs the right tools in place to effectively drive the business forward.
Investing in Data Science
Prior to the modern stack in 2012, most analyses in the business either ran off of Excel or required an engineer to do SQL. That all changed when one of the first data scientists brought on Fivetran, Redshift and Looker. “Anyone that was at the organization when it had its inefficient SQL setup gives the new stack a 10/10,” says Radiyat.
Since the initial implementation, Muskaj has added the Heroku connector to ingest data from its Clover app: “I don’t know what alternative approach we could have taken to get our Heroku data into Looker. With Fivetran I can just create a user, pop in the permissions and boom. It made it so much easier. Our general experience with Fivetran is that it is easy. We would need an additional employee to do the work of Fivetran.”
The business also makes heavy use of the Uploads function. “We love Uploads because it gives people the ability to upload documents and have them accessible in Looker. It is a very useful feature,” Muskaj says.
Scaling Up With AWS
When DonorsChoose rebuilt its web system on AWS, it migrated its PostgreSQL production database to AWS Relational Database Service. “It was so easy to set up the Fivetran connector to point to the new destination,” explains Muskaj. “We just updated the permission information and now our team can leverage AWS to address the scaling issues and high availability issues that we were having previously.” DonorsChoose also uses the S3 connector to load files from S3 into Redshift.
Because DonorsChoose doesn’t have to write scripts for its PostgreSQL or Heroku pipelines, it has time to model the data in Looker for stakeholders. “Our dashboards update data at whatever interval we choose. We can focus much more on sharing insights with the business instead of ensuring that we’re getting the correct data,” Radiyat says. “We’re able to make any new fields that we add to the Heroku database easily accessible in Looker.”
Visualizing Data to Make Business Decisions
The business is especially interested in the data from its recently released Round Up for Schools app, which merchants can add to their Clover point-of-sale devices for free. It enables customers to round up their purchase to make a donation to a local school. “This can be really powerful for us. The dashboard tells us how the app is doing and how many users we’re getting. For example, it allows us to visualize merchants that were once performing really well but have slowed down. We can reach out to them and see what is going on.”
Ultimately, the goal of the business is self-service and efficiency. “Generating reports used to involve copying and pasting data into Excel, copying them into another Excel sheet and compiling them into a report. It was a weekly report and often stale,” says Muskaj. “Now people can click one button and see those results. Our active BI users continue to grow with this stack in place and we have 100 active Looker users in our organization of 167.”
About Fivetran: Shaped by the real-world needs of data analysts, Fivetran technology is the smartest, fastest way to replicate your applications, databases, events and files into a high-performance cloud warehouse. Fivetran connectors deploy in minutes, require zero maintenance, and automatically adjust to source changes — so your data team can stop worrying about engineering and focus on driving insights.
About Amazon Redshift: Amazon Redshift is a fast, fully managed, petabyte-scale data warehouse solution that makes it simple and cost-effective to efficiently analyze all of your data using your existing business intelligence tools.
About Looker: Looker is a modern platform for data that offers data analytics and business insights to every department at scale, and easily integrates into applications to deliver data directly into the decision-making process.