Finding Metrics that matter and using them safely
summary
Claims a metric program must start from an outcome and work back to measures, using outcome → decision → insight → measure; shows concrete charts (cycle time jitter plots, net‑flow open/closed bars, throughput by month) and recommends tools like probabilistic forecasting and ActionableAgile. Highlights a surprising benchmark: flow efficiency ≈ 5% for many teams and good targets near 20–25%. Proposes a single balanced dashboard tying deployment frequency to value, warns metrics get an expiration date, and gives concrete steps to make metrics safer.
AI-written summary · about
chapters (34)
- 0:00Intro
- 1:02Too many metrics
- 1:45Baseline metrics
- 4:59Finding the right metrics
- 6:09Use the right metrics
- 6:56Odom
- 9:41Outcome
- 15:09Silhouette Test
- 16:29Balanced Metrics
- 18:34Finding Balanced Metrics
- 20:47Quality Metrics
- 23:34Customer Satisfaction
- 24:50Responsiveness
- 29:03Vanity
- 30:49Deployment Frequency
- 31:25Open vs Closed
- 32:32Team Happiness
- 35:08One Visual Space
- 36:02Flow Efficiency
- 38:37Summary
- 39:07Trends
- 40:46How to Measure Anything
- 41:42Missteps
- 42:06Over optimizing
- 42:19Awareness
- 45:48Carmelo Anthony
- 47:10Troy McGinnis
- 48:36blindly pursuing targets
- 49:46numerical targets
- 50:00dark side of metrics
- 50:17remember what you measure
- 50:49measure me in illogical ways
- 51:50exploring the unhappy path
- 53:55Questions
related talks
28mMetrics that Matter
27mMetrics that Matter - Moving from Easy to Impactful
47mWhat's Measured Improves: Metrics that Matter
23mYou Get What You Measure—Why Metrics Are Important
26mMetrics and Modeling – Helping Teams See How to Improve
39mTest Metrics – a thread or a threat
06mDelivery Metrics: the good, the bad, and the totally ridiculous
05mMetrics That Matter