Mind Your P’s and Queues
summary
Agile prioritization orders when work starts but value is realized only when items finish, so priority often fails if queues (WIP) explode. Uses Monte Carlo simulation and Kanban-style boards to show an 18‑feature backlog with WIP=9 can make priority #1 less likely to finish than nine lower priorities. Critiques weighted shortest job first and shows common Monte Carlo assumptions (one thing at a time, strict priority) are false. Main actionable claim: measure and cap queue size to make forecasts meaningful. Monte Carlo, Little’s Law
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:04Origin of the expression
- 4:17P-words in agile
- 5:43Prioritization problems
- 8:51Value realization in agile
- 10:37Flaw in prioritization
- 12:00Queues in agile workflows
- 15:52Release timeline questions
- 17:37Monte Carlo limitations
- 20:45Prioritization waste in WIP
- 22:12WIP impact on planning
- 22:34Future AI talk teaser
- 23:38Question and answer session begins
related talks
27mWhat does "high priority" mean? The secret to happy queues
30mHow to Visualize Impacts to Your Workflow & Metrics
35m10 Tips for Product Prioritizations
46mEstimates and complexity: a tale in 42 and a half minutes
47mMinimum Viable Process
23mThe Shape of Uncertainty
10mPractical tech-debt prioritization
42mHow Flow Works and other curiosities