Being Right Starts By Knowing You're Wrong
summary
Argues that companies get the most value from explicitly measuring error, showing concrete metrics and examples: expected 1,000 units vs 670 (330 units error), expected $500k vs $1.5M (67% error), median absolute percent error ~28.4 with IQR 21–47 and mean APE 30.17 across quarters. Recommends simple measures—mean absolute error, mean absolute percent error—applied to demand planning, sales, marketing and people analytics; time‑series tracking and basic tools (even Excel is enough) let teams learn and improve.
AI-written summary · about
chapters (9)
- 0:00Introduction and welcome
- 1:30Talk overview and cultural challenges
- 4:20Defining error and its importance
- 9:16Measuring error simply and effectively
- 18:45Error as a learning opportunity
- 25:50Business as a complex neural network
- 26:33Applying error measurement in business
- 35:01Cultural change and analytics mindset
- 35:45Q&A session
related talks
32mStatistics for Engineers
34mTen Tips to Leverage Data Better and Make Better Products
32mIntroducing Reality into Data Metrics
18mHow Data Scientists Broke A/B Testing (And How We Can Fix It)
32mThe Subtle Art of Lying with Statistics
27mThe Subtle Art of Lying with Statistics
05mPerception and bias and metrics, oh my!
15mMachine Learning: 5 Additional Metrics to Avoid High‑Performing but Useless Models