Software Engineering Radio - the podcast for professional software developers SE-Radio Episode 360: Pete Koomen on A/B Testing
Mar 13, 2019
Pete Koomen, Co-founder and CTO at Optimizely, dives deep into A/B testing, sharing insights from running over 200,000 tests. He reveals how structured experimentation can elevate product innovation, highlighting the importance of hypotheses and success metrics. The conversation also addresses the challenges of testing, the significance of understanding user behavior, and the ethical considerations in experimentation. Koomen emphasizes fostering a culture of testing within organizations to drive data-driven decision-making and enhance user experience.
AI Snips
Chapters
Transcript
Episode notes
A/B Testing Basics
- A/B testing runs experiments by showing users different variations to determine which performs best. - This method improves decision-making based on data, not opinions.
Run Tests Without Code Changes
- Use visual editors and client-side libraries to run A/B tests without changing website code. - Avoid engineering team dependency for minor experiment changes to increase agility.
Define Clear Success Metrics
- Define success metrics aligned with your organization before starting experiments. - Tailor success criteria like donations or email signups to your specific product goals.