Growth Hacking - Product Analytics at Scale using R and RStudio
summary
Shows how R and RStudio tools produce product-retention KPIs at enterprise scale using tidyverse plus plotly and formattable, with Python interop via reticulate. Starts from raw log data, computes user age/cohort with two to three lines of dplyr, and demonstrates an example dataset of 13,000 users falling off over months. Converts ggplots to interactive htmlwidgets, surfaces cohort sizes and industry splits, and maps retention to survival methods (Kaplan‑Meier) so the outputs are reproducible, shareable, and deployable.
AI-written summary · about
chapters (17)
- 0:00Intro
- 1:48Personal Background - Andrew Mangano
- 3:35Salesforce Customer 360
- 4:48The Customer Journey Adoption - Retention - Expansion
- 5:46Retention is key to Growth Hacking Focus on keeping the users you have
- 6:59Challenges of retention analytics Getting from data to insight
- 8:52Base R doesn't compute monthly intervals
- 10:41Use ggplot to build a retention curve
- 12:13Use Plotly for interactive charts to easily see differences
- 12:34R can handle complex data!
- 14:06Clean up the output
- 14:22Stack detail using pipes % %
- 14:55Using knitr and formattable, create a readable table
- 15:26RStudio can handle complexity
- 16:27Export to an html widget to host and share with end users
- 18:18Survival Analysis can add more science to your Data Science
- 19:43Values Drive Value Doing well and doing good
related talks
53mReal-Time Big Data Analytics with R
46mElegant R: Tips from a Software Developer
34mBuilding Data Driven Products with Ruby
19mA Journey Into Data Science: From an Idea to a Viable Product
15mAdvocating for Automation: Adapting Current Tools in Environmental Science with R
48mHuman data interactions: experiments in how humans analyze data
24mStorytelling With Data
29mGrammars of Data: lessons from ~20 years of the tidyverse