Story Points Suck!

Transcript

1. Exploring a unique approach to estimation Managing Consultant Telstra Purple
2. THE HISTORY
3. …using them [story points] to predict “when we’ll be done” is at best a weak idea
4. PROBLEMS WITH STORY POINTS
5. 7:20AM Story Point
6. How long does it take to get to work? EXPERIMENT 1st March 2nd March 7:00AM 8:10AM 7:00AM 7:30AM 3rd March 7:00AM 7:20AM
7. Plans based on average are wrong on average
9. PRINCIPLES OF FORECASTING Reforecast with new information Think probabilistically, not deterministically
11. Scatterplot Cycle Time (Days)
12. DON’T ESTIMATE STORIES
13. WORK IN PROGRESS
14. Monte Carlo Simulation
15. EXPERIMENT
16. EXPERIMENT
17. 1 4 10,000x BASIC
18. When will all the work be 1 1st Feb 2nd Feb 3rd Feb 4th Feb 0 2 1
19. When will be
20. How stories can be
21. START FORECASTING in just Four weeks
22. LITTLE’S LAW AVERAGE CYCLE TIME = AVERAGE WIP / AVERAGE THROUGPUT
23. Use the assumptions of to verify stability
24. CFD
25. Average Arrival Rate Average Departure Rate matches
26. Finish all work that is started
27. Work in Progress should be constant
28. Work in Progress should be constant Average Age of
29. USE consistent UNITS
30. Think probabilistically
32. scatterplot
33. stories with
34. Completion Predict with Monte Carlo Simulation
35. Ensure stability with assumptions of
36. IMPACTS Tomorrows PREDICTABILITY Today’s actions
37. Thank you