Why data science and UX research should be best friends
summary
Explains concrete collaborations: a booking-site model that predicts host acceptance and drives a 10% increase in conversions, a deep learning model that predicts element tapability at ~90% of human judgment, a music-service diary study that uncovers a user belief causing skipped-ad limits, and a bank’s tagger that spits 20–30 tags and increases time on task (a failure that needed UX tightening). Outlines the data pipeline from wrangling to evaluation and shows using models to predict churn and to prioritize features.
AI-written summary · about
chapters (11)
- 0:00The importance of data
- 2:12Defining UX and data
- 3:32The data science pipeline
- 6:02Technical data processes
- 11:21Data storytelling
- 12:57Case studies of collaboration
- 17:11Spotify's insights
- 20:39Lessons from failure
- 22:48Holistic user understanding
- 25:44Actionable next steps
- 30:07Conclusion and resources
related talks
33mUX Research and Design: Crucial to development
43mUX + Data = Love
12mStart Using UX as a Weapon
44mHow to Improve UX with Machine Learning: Insights and learnings until now
04mDBTB INT Hunter Whitney r
24mLeveraging Automated Testing Data to Enhance User Experience
21m🎙️UX+DEV SUMMIT 2018 - Pedro Marques
46mFundamentals of UX Design and Usability