Transparency USA is a non-profit organization whose mission is to provide “clear, accurate, easy-to-understand information about the money in state politics.” Originally, Transparency Texas, in early 2020, expanded from displaying only Texas data to supporting eight additional “Swing States” in less than five months.

We have been a partner with them since day one to help manage multiple pieces of the project:

Data Processing
Dealing with the government is something most people dread, but it’s our favorite part about this project! We built a custom Node.js framework that helps us procure data from each government entity, import the data, clean it up, and export it into a standard format for further processing.
Data Dedupe
Manual data entry is pretty messy. It’s ripe with errors, misspellings, partial information, and special characters. To help improve data quality, we built a tool that allows our client to manually process and review over 25 million records. There are also various modules that allow them to integrate with 3rd parties, such as Ballotpedia and dedupe duplicate records.
Web Application
Government websites are hard to use. They are usually slow, ugly, and make it impossible to find the information you need. After obtaining data and cleaning it up, we wanted a web application that was easy for your everyday American citizen to utilize. The web application we built makes it easy to consume information in a consistent format across nine states! We track almost $10 billion dollars worth of transactions and display it on pages that load in milliseconds (not minutes).
Ballotpedia Widget
Once you have the best data in town, you want to partner with another “best in business” organization. We worked with Ballotpedia and Transparency USA to build a custom integration between their two data sets to enable bi-directional sharing. This enabled our client’s data to show up on millions of page views a month! It also allowed Ballotpedia to slice and dice campaign finance data and answer questions that were never possible before.