Cascade was a no-code data logic and analysis tool, allowing business analysts and other non-technical teams to translate raw data into clear insights. Analysts could build analysis workflows pulling tables from databases, apps or flat files and producing insights and visualizations they could share with others. Each workflow is repeatable, so when any underlying data changes, the results can automatically update.
In its heyday, Cascade was used by a scooter company to manage repair and dispatcher data, by a contracting firm to analyze customer and billing data, and by HR firms to analyze and translate employee census files. We knew that each use case would be unique and specific to the business encountering it, but that they all boiled down to business teams needing to automate data logic.
The app split into two sections, a Workflow Canvas and a Data App Interface. The first and primary section was a no-code free-form canvas, allowing analysts to assemble data logic in the way that they need. Workflows are made out of a set of 40+ tools, with each tool transforming data in some way. Tools included importing tables, joining, calculating or aggregating datasets, or visualizing results.
One of the primary tools in our arsenal was what we called Edit Columns, which contained a complete library of Excel-like formulas to run analysis and create calculations. That library allowed teams with only Excel knowledge to create repeatable, automated workflows that could handle the large datasets common in today’s enterprises.
Once a workflow is built, tables can be visualized and presented on the second major section of the app, the Data App Builder. This is where analysts can arrange inputs, tables and charts to create interactive mini-applications that can be shared with others.
We found that in many cases, Cascade replaced Excel or an unautomated, spreadsheet-based solution. While spreadsheets enjoy near-universal adoption, many companies pay a silent tax as their employees repeat the same operations again and again to get updated results. Cascade eliminated that repetition. In many other cases, alternatives included older automation platforms like Alteryx or custom-built solutions in code. In these cases, we found that business teams valued the ability to manage data logic themselves or migrate to a more collaborative, modern platform.
All in all, we were delighted to see the amount of engagement that occurred when analysts committed to the platform. In some cases, we became an IDE of sorts for analysts, with some spending most of their working weeks inside the app. The general-purpose nature of Cascade allowed our customers to use it for any data task that came their way, but it also raised the bar for what the product needed to handle and how well it could be sold. We were progressing quickly enough to build a good business, but not quickly enough to justify venture backing or our time as founders.
