What data and design can learn from each other
summary
Argues that data awareness — treating design thinking and the scientific method as equal — prevents wasted wrangling and produces clearer products. Shows a Seattle collisions EDA done in Python in a Jupyter notebook where geocoding (attempted with regis) fails and the student manually bins lat/long; proposes collecting neighborhood in the UI or adding a data persona upstream. Uses matplotlib and seaborn examples, Anscombe’s quartet and Nightingale to justify changing opacity, line weight, ticks and fonts, and using a dash for nulls after user testing.
AI-written summary · about
chapters (26)
- 0:00Introduction and welcome
- 1:12Why this topic matters
- 2:59Defining design and data
- 4:02Scientific method vs design thinking
- 5:49Data awareness vs data driven
- 8:37Applying design to data collection
- 11:53Data aware design opportunities
- 13:19Designing for downstream users
- 16:08What is data visualization
- 17:10Florence Nightingale’s coxcomb
- 18:57Anscombe’s quartet explained
- 20:42Data visualization in Python
- 21:45Beauty in data visualization
- 22:26Audience and design considerations
- 23:30Improving graphs with code
- 24:32Data visualization in UI design
- 25:57Handling null values in UX
- 28:05Key takeaways summary
- 28:47Actionable takeaways for roles
- 30:51Q&A session begins
- 31:12Question about Starbucks role
- 31:58Recommended books and resources
- 36:57Designer’s journey into data science
- 39:06Entering analytics field advice
- 40:50Dash vs blank in UX testing
- 42:17Closing remarks and survey
related talks
27mVisual Design - Do not let Data be your cul-de-sac!
37m“Design Systems for Data Products”
01h 38mEffective Data Visualization
02h 31mTech Tutorial: Data visualization in Python
06mData + Empathy: How to Approach UX Decisions in Web3
24mDesign and Development between Art and Science
31mData Storytelling through Visualization
47mData, Design, Code