Building a Flexible, Scaleable Self-Serve Reporting System with Shiny
summary
A Shiny app called betterup insights pulls live data from the company database, renders professional charts, and injects metrics and images into preformatted Google Slide templates, cutting a reported 18 hours per business review and supporting a baseline ROI of about $1.2M/year. The build uses a module-first Rhino framework, R6 objects with Garo triggers to control reactivity and enforce privacy thresholds, and reticulate to reuse Python Google API code. The system ships reusable templates and wins Posit procurement.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 2:00How long does it take
- 2:41Two crucial questions
- 3:10Current approaches
- 5:23User interviews
- 5:54Introducing Betterup Insights
- 7:34Build Architecture
- 9:25Benefits
- 10:50Surfing
- 11:37Identify a core capability
- 12:39Build your team
- 14:35Build a data product
- 16:40Measure impact
- 17:36Screenshots
- 18:13Thank you
- 19:02Question
related talks
18mBuilding an AB testing analytics system with R and Shiny
35mBuild Simple and Scalable Apps with Shiny
17mSaving 1,000 hours with RStudio: selling R in your workplace
09mPublishing Customer Facing Products with RStudio Connect
12mDashboard-Builder: Building Shiny Apps without writing any code
15mR Markdown + RStudio Connect + R Shiny
27mShiny: A GUI for the Tidyverse
13mIntroducing Rhino: Shiny application framework for enterprise