Little's Law: Work in Progress, Throughput, and Cycle Time
October 29, 2026
Little's Law is a simple, powerful relationship for any stable system: the average number of items in the system (L, or WIP) equals the average arrival or completion rate (λ, or throughput) multiplied by the average time each item spends in the system (W, or cycle time). L = λ × W. Rearranged: cycle time = WIP ÷ throughput.
A team completes 20 items a week and has an average of 60 items in progress. The average cycle time is 60 ÷ 20 = 3 weeks. That is the average time from starting work to finishing it.
Skip the math and get your answer instantly:
Open the Average Calculator →Using It to Reduce Lead Time
If throughput stays at 20 per week and the team limits WIP to 30, the average cycle time falls to 30 ÷ 20 = 1.5 weeks. Work gets done faster, and feedback comes sooner, without anyone working harder. This is the logic behind WIP limits in kanban. The law doesn't say that reducing WIP always keeps throughput constant; if WIP is so low that people are idle, throughput can fall. There is a sweet spot where throughput is maintained.
| System | Throughput | Cycle time | Average WIP (L = rate x time) |
|---|---|---|---|
| Software team | 20 items/week | 3 weeks | 60 items |
| Same team with WIP limit | 20 items/week | 1.5 weeks | 30 items |
| Coffee shop | 30 customers/hour | 4 minutes (0.0667 h) | 2 customers inside being served |
| Hospital ward | 5 admissions/day | 4 days | 20 occupied beds |
Other Applications
A hospital ward admitting 5 patients a day with an average stay of 4 days has about 20 patients at any time, which tells you how many beds are needed (plus margin for variability). A website with 50 visitors arriving per minute that stay 3 minutes on average has about 150 concurrent users, which helps size servers. A store with 100 customers per hour and 12 minutes in the store has 20 customers inside. Wherever you know two of the three numbers, you can find the third.
Conditions and Pitfalls
The system must be stable over the measurement period: arrivals equal departures on average, and no big growth or decline in the backlog. Use consistent units and consistent definitions of when an item starts and finishes. The law deals with averages and doesn't depend on the arrival distribution or the order of service. Do not confuse cycle time (time in the process) with lead time (from request to delivery, including waiting before work begins), although the law applies to either when the system boundaries are defined consistently.
Practical Steps
Measure throughput and WIP over a few weeks, compute the implied cycle time, and compare it with actual measurements as a data-quality check. Set WIP limits, watch what happens to the cycle time, and adjust. Reducing the variability of work items also shortens queues.
Frequently Asked Questions
What is Little's Law?
The average number of items in a system equals throughput times the average time in the system.
How do I compute cycle time?
WIP divided by throughput.
Why do WIP limits shorten cycle time?
For constant throughput, fewer items in progress means each spends less time in the system.
Does it require a stable system?
Yes, on average over the period.
Does it depend on the arrival pattern?
No, it holds for averages regardless of the distribution.