Building a Marketing Analytics Web App with Python
summary
Claims a full replacement for a proprietary marketing analytics platform by wiring Shopify data into a Python stack and shipping a dashboard plus downloadable HTML reports. Uses Singer/Meltano for extracts, DuckDB as the OLAP layer, pandas/polars and Altair for transforms and plots, and DataPane to render apps and reports. Demonstration runs on ~1M cells, the app logic is ~300 lines of code, schedules ETL jobs, and the author says teams can deliver an end-to-end product in under a day using data pane packages.
AI-written summary · about
chapters (22)
- 0:00Replacing SAS with Python
- 1:05What is DataPane
- 3:42Python for non-technical users
- 6:15Proprietary platform limits
- 8:00Data apps vs full products
- 10:10Python vs off-the-shelf analytics
- 11:14Demo of web app
- 14:25App components breakdown
- 15:09Efficient data extraction
- 16:32Choosing data storage
- 17:58Data transformation in Python
- 18:40Scheduling data tasks
- 19:22Reports with DataPane
- 20:27Core analysis libraries
- 21:31Pre-rendering for performance
- 22:12Automating insights
- 23:20Code walkthrough
- 26:10Realistic for non-experts
- 26:52DataPane deployment packages
- 28:57TL;DR and alternatives
- 30:25Q&A: Authentication and datasets
- 35:31Closing remarks
related talks
31mBuilding Powerful Marketing Apps
54mHow to build stunning Data Science Web applications in Python
40mOpen Data Dashboards & Python Web Scraping
41mDash: data exploration web apps in pure Python
02h 14mData Applications with Bokeh
01h 07mBig_Data_Meets_Digital_Advertising
25mHow Snowflake Powers Marketing Analytics
12mAnalytics architecture strategy for intelligent experiences