Value Stream Metrics
This domain tests whether you can choose flow metrics over resource metrics and compute them correctly, including the arithmetic of Little's Law and the difference between averages and distributions.
Lead time and cycle time
Lead time runs from the customer's request to delivery of value, including all the waiting. Cycle time runs from the moment work starts to the moment it finishes. Lead time is the customer's experience; cycle time is the team's. Reporting one while calling it the other is a standard exam trap.
The flow metrics
- Throughput: items completed per unit time.
- WIP: items started but not finished.
- Flow efficiency: value-added time divided by total lead time.
- %C&A: the share of items arriving complete and accurate, exposing hidden rework.
- Flow time, flow load, flow velocity, flow efficiency, and flow distribution (Kersten's Flow Framework) give a product-level view.
- DORA's four keys - lead time for changes, deployment frequency, change failure rate, time to restore service - connect flow to delivery performance.
Little's Law
Average time in system = average WIP ÷ throughput. It holds for any stable system and is the fastest way to sanity-check a claim: if you want a two-week lead time at six items per week of throughput, WIP must stay near twelve.
Worked example
A team reports 24 items in progress and 6 completed per week, so average lead time is 24 ÷ 6 = 4 weeks. Leadership wants two weeks. The reasoning: throughput is a consequence of the constraint, so the lever is WIP - cap it at 12 and the same throughput yields two weeks. Also compute distributions: if the 85th percentile is 9 weeks, some customers still wait far longer than the average suggests.
Traps that cost marks
- Treating utilization near 100% as efficiency; in queueing terms it produces runaway waiting.
- Using story points or lines of code as flow metrics.
- Reporting averages only, hiding a long tail.
- Measuring individuals rather than the stream.