Weighted Shortest Job First (WSJF): Prioritizing for Economic Impact

THE MENTAL MODEL: Weighted Shortest Job First (WSJF) is a prioritization framework that helps you sequence work to maximize economic benefit over time. Instead of asking "What is the most valuable feature?", it asks, "What is the most valuable feature we can deliver quickly?" It's a model for continuous flow, not for creating a static, ranked list.

HOW IT WORKS: WSJF is calculated with a simple formula: WSJF = Cost of Delay / Job Duration. The highest score wins and gets done next. Both inputs are relative estimates, not absolute numbers.

Cost of Delay is the sum of three factors, each estimated relatively (e.g., using a Fibonacci sequence like 1, 2, 3, 5, 8): first, User-Business Value; second, Time Criticality (does value decay over time?); and third, Risk Reduction or Opportunity Enablement (does this unlock future value?).

Job Duration is a relative estimate of the effort required to complete the job. Using a relative size prevents teams from getting bogged down in promising specific timelines.

By dividing value by effort, the model favors smaller, valuable items, getting them delivered sooner. It also inherently ignores sunk costs, focusing only on the economics of what to do next.

WHEN TO USE IT: Use WSJF in flow-based systems like SAFe to prioritize a backlog of features or epics. It is most effective when you have more competing ideas than you have capacity to build, and you need an objective framework to facilitate trade-off conversations among stakeholders.

WHEN NOT TO USE IT: WSJF is overkill for simple backlogs or when priorities are obvious. It's also not meant for work with fixed, non-negotiable deadlines like regulatory compliance; that work typically bypasses this kind of economic sequencing. If you spend more time debating the WSJF inputs than it takes to build a small feature, you are using it incorrectly.

ONE CANONICAL EXAMPLE: Imagine two features. Feature A has a high Cost of Delay (Value=8, Time=5, Opportunity=5) for a total of 18. Its Job Duration is large, estimated at 8. Its WSJF score is 18 / 8 = 2.25. Feature B has a slightly lower Cost of Delay (Value=5, Time=8, Opportunity=3) for a total of 16. Its Job Duration is very small, estimated at 2. Its WSJF score is 16 / 2 = 8.0.

Despite Feature A having a higher raw Cost of Delay, Feature B should be done first because its WSJF score is significantly higher. This delivers economic value to the business much faster.