Download as PDF, PPTX
Uploaded byAgile Montréal
Actionable Agile Metrics for Predictability - Daniel Vacanti
The document discusses using cycle time and scatterplots to measure and visualize workflow to better predict completion dates for work items. It introduces the concept of thinking probabilistically rather than deterministically when forecasting, since multiple outcomes are possible. Specific metrics like the 50th, 85th and 95th percentiles of a cycle time scatterplot can provide forecasts of when a given percentage of items may complete. Factors like work in progress, blockers and dependencies affect cycle times as well. The document promotes tracking start and end dates to measure flow and cycle times as fundamental metrics for gaining predictability.
More Related Content
![Metrics at Every (Flight) Level [2020 Agile Kanban Istanbul FlowConf]](https://cdn.slidesharecdn.com/ss_thumbnails/metricsateveryflightlevel2020agilekanbanflowconf-201208160904-thumbnail.jpg?width=640&height=640&fit=bounds)
Similar to Actionable Agile Metrics for Predictability - Daniel Vacanti
Actionable Agile Metrics for Predictability - Daniel Vacanti
- 1. Daniel S. Vacanti ActionableAgileTM daniel@actionableagile.com @danvacanti ActionableAgile Metrics for Predictability
- 2. First, let meset the record straight…
- 3. “When will itbe done?”
- 4. “When will itbe done?” Date (number of days) Elapsed Time
- 5. “When will itbe done?” Date (number of days) Story Points / Velocity
- 6. Stop me ifyou’ve heard this one before…
- 7. “Relative Complexity isthe best predictor of how long it takes an item to complete”
- 9. Why do weeven bother with Story Points?*
- 10. I’m going tosuggest something radical…
- 11. As an example: Howlong does it take you to get to work in the morning?
- 12.
- 13.
- 14. 4 4 2 ActiveDone Testing Done DoneActive DevelopmentReady Analysis 3 Here’s a spot!!
- 15. Try an experimentfor me…
- 16.
- 21. Try the samething for your process
- 22. Backlog Analysis DevelopTest Deployed Start Timer Stop Timer
- 23. This chart iscalled a (Cycle Time) Scatterplot
- 24. If you tracknothing else, track the date that an item starts and the date that an item completes (for all work items)
- 25. That will giveyou a measure of the flow metric of Cycle Time
- 26. Cycle Time isthe amount of elapsed time it takes for a given work item to complete
- 27. “When will itbe done?” for a single item is best answered by Cycle Time
- 28.
- 29. How does generatinga chart like this help us answer “When will it be done”?
- 30. How do wemake sense of this randomness?
- 31. Your process is“random”. Therefore, you can’t think deterministically. You need to think probabilistically.
- 32. What does itmean to think probabilistically?
- 33. Let’s try anotherexperiment…
- 34. Thinking probabilistically means acknowledgingthat there is more than one possible future outcome
- 35. How many peoplecan we “expect” to be standing after 3 flips? 12.5% Does that mean 12.5% was the only possible outcome?
- 36. Backlog Analysis DevelopTest Deployed 2 Start Timer Stop Timer
- 37. Backlog Design DevelopTest Deployed 2 Exactly this dot? Exactly this dot? Or Exactly this dot? Will Item #3 end up as…
- 38. Backlog Design DevelopTest Deployed 2 There is more than one possible outcome for Item #3 while it is sitting in the backlog
- 39. What are thepossible outcomes?
- 40. 40 How do wemake sense of this “randomness”?
- 41. 41 50th Percentile =6 days or less
- 42. 42 85th Percentile =15 days or less
- 43. 43 95th Percentile =22 days or less
- 44. 44 50th Percentile =6 days or less 95th Percentile = 22 days or less 85th Percentile = 15 days or less Scatterplot Percentiles (making a forecast)
- 46. 85th Percentile =33 days or less
- 47. 85th Percentile =13 days or less
- 50.
- 51.
- 53. Avg Cycle Time= Avg WIP Avg Throughput
- 54.
- 55. Poor Pull Policies, Blockers, ExternalDependencies, …
- 56.
- 57. What does a19th century Yorkshire cotton industrialist have to do with the Manhattan Project? For next time…
- 58.
- 59. All charts createdby: Thank-you!