Lesson 2 · Delivery
Flow, WIP & cycle time
To finish faster, start less. The arithmetic is not optional.
Delivery is a flow problem. Throughput is how many items finish per unit of time, cycle time is how long one item takes from start to finish, and work in progress (WIP) is how many are underway. Little's law ties them together: average cycle time = WIP ÷ throughput. If throughput is fixed, the only way to finish things sooner is to start fewer of them.
That is why WIP limits work. Ten items in progress with a throughput of five a week means a two-week average cycle time; limit WIP to five and it halves, with the same team and the same effort. High WIP also hides problems, multiplies context switching and delays feedback. Stop starting, start finishing.
Every system has a bottleneck that sets the throughput of the whole — Goldratt's theory of constraints. Improving anything else does not help: faster coding only piles work up in front of a slow review or release step. Find where work waits longest and fix that, then look for the next constraint. Measure flow efficiency, active time over elapsed time, and you will usually find that work waits far longer than it is worked on.
Picture it
A team feels busy but ships slowly: 24 tickets in progress and about 6 finished a week, a four-week average cycle time. The board shows 11 tickets waiting for review. The manager sets a WIP limit of 12 and one rule: review before starting anything new. Throughput stays near 6, cycle time falls to two weeks, and customers see fixes twice as fast.
Key formulas & frameworks
- Little's law
- Cycle time = WIP ÷ Throughput
- Flow efficiency
- Active time ÷ Total elapsed time
- Theory of constraints
- The bottleneck sets the throughput of the whole system
- WIP limit
- Stop starting, start finishing
Practice
Score: 0 / 5
Up nextCapacity & utilization
Next lesson