Cycle time is the internal time a process takes to complete one unit; lead time is the total time a customer waits from placing an order to receiving it. Cycle time is measured at a station in seconds or minutes. Lead time runs the whole journey and is usually measured in days, because it includes all the waiting between operations.
These are the two most-confused numbers on a plant floor, and the confusion is expensive: teams quote a two-hour cycle time to a customer who then waits two weeks, or they attack cycle time to fix a lead-time complaint and wonder why nothing improves. One diagram fixes the mix-up for good, cycle time is a small segment that lives inside lead time. This guide draws that picture, explains how the two connect, and shows which one to measure for which problem.
What Is the Difference Between Lead Time and Cycle Time?
The difference is scope and point of view. Cycle time is internal and narrow, it measures work at a single process, from the start of one unit to the start of the next, and it is what your equipment and operators control directly. Lead time is customer-facing and wide, it measures everything the customer experiences, from the moment they place an order to the moment it arrives, including order entry, planning, queue time, processing, inspection, packing, and shipping.
The practical tell: cycle time is what your process does; lead time is what your customer waits. A filler with a 45-second cycle time can sit inside a lead time of ten days, because the case it fills spends most of those ten days waiting, in a queue before the filler, in a hold after it, in staging for a truck. Cycle time never sees that waiting. Lead time is almost nothing but that waiting.
How Are Lead Time and Cycle Time Related?
Lead time contains cycle time plus all the waiting between and around the processing steps. Add up every station's cycle time and you get the total work content of an order; the gap between that sum and the actual lead time is pure waiting, queues, holds, transport, and staging. In most plants that gap is enormous, because processing is only 5–10% of lead time and waiting is the rest.
Little's Law ties the two together at the system level: lead time equals work-in-process divided by throughput. So lead time is not set by how fast one station cycles; it is set by how much WIP is in the system relative to how fast the system finishes units. You can cut every cycle time in the plant and barely move lead time if the WIP piles between operations stay the same. That is why lead-time complaints almost never get solved by speeding up a machine, the fix lives in the waiting, covered in our lead time reduction guide and grounded in the math of Little's Law.
| Cycle time | Lead time | |
|---|---|---|
| What it measures | Work at one process, per unit | Order-to-delivery, the whole journey |
| Point of view | Internal (what the process does) | Customer (what the customer waits) |
| Typical units | Seconds or minutes | Hours, days, or weeks |
| Includes waiting? | No, only the active cycle | Yes, waiting is most of it |
| You improve it by | Faster methods, less variation at the station | Smaller batches, less WIP, shorter queues |
| Best tool | Cycle-time study / machine data | Value-stream map, WIP cap |
Which One Should You Measure?
Measure both, because they answer different questions and neither substitutes for the other. Reach for cycle time when the question is about a station: is this machine keeping pace with takt where is the bottleneck, is one operation slower than the others? Reach for lead time when the question is about the customer: why are orders late, how long does an order really take, can we promise a shorter delivery? Using the wrong one is how improvement effort lands in the wrong place.
A worked example makes the split concrete. A shop quotes a customer "two hours" because the routing sums to two hours of cycle time across five operations. The customer's order actually ships in nine days. Where did nine days come from? The two hours of work sat inside nine days of queue: the job waited two days for the first machine, a day between each of the next four operations, and two days in final inspection and shipping. Cycle time is honest, two hours of work is correct. Lead time is what the customer feels, nine days. Quoting cycle time as if it were lead time is a promise the plant cannot keep.
The example also shows why the two metrics improve on different clocks. If the shop wanted to shave the two hours of cycle time, it would tune methods, reduce variation at the slow station, or add a faster machine, real work with real cost that would move lead time from nine days to eight days and eleven hours. If the shop wanted to shave the nine days, it would shrink batches, cap the WIP sitting between operations, and pull the two-day inspection queue apart, cheaper moves that could take lead time from nine days to four without touching a single cycle time. Same order, two entirely different improvement programs, chosen by which metric you were actually complaining about.
Lead time vs. cycle time: the reference numbers
The relationship rests on a few well-established facts:
- Little's Law: lead time = WIP ÷ throughput the queuing identity that governs how the two metrics connect. A system holding 500 units and finishing 100/day carries a 5-day lead time regardless of any single cycle time.
- Processing is typically 5–10% of manufacturing lead time; the remaining 90%-plus is waiting. Summed cycle times therefore understate lead time by roughly a factor of ten.
- U.S. manufacturing runs near 75.7–76.3% capacity utilization per the Federal Reserve's G.17 release (Federal Reserve, Industrial Production and Capacity Utilization); as utilization climbs toward 100%, queue time and therefore lead time rise sharply while cycle time stays flat.
How Do You Keep the Two Straight in Daily Use?
Keep them straight with a short habit: name the point of view before you name the number. When someone on the floor says "time," ask whose clock they mean, the station's or the customer's. The routine below stops the two from blurring in meetings and quotes:
- Ask "whose time?" first. If the answer is a machine or a station, it is cycle time. If the answer is the customer waiting for an order, it is lead time. This one question resolves most arguments before they start.
- Check the units. Seconds or minutes almost always means cycle time. Days or weeks almost always means lead time. A "time" quoted in the wrong units for the context is usually the wrong metric.
- Never quote a delivery date from summed cycle times. Add the queue. Order-to-ship promises come from lead time, which includes the waiting your routing does not.
- Attack the right metric for the complaint. "This station is slow" is a cycle-time problem. "Our orders are late" is a lead-time problem. Fixing one when the customer meant the other wastes the effort.
- Watch them on separate charts. A cycle-time trend per station and a lead-time trend per order both belong on the board of manufacturing KPIs. When cycle times are flat but lead time is climbing, you have a queue problem, not a machine problem.
Where Does the Confusion Come From?
It comes from the word "time" doing double duty, and from people quoting the number they can measure easily instead of the number the customer asked about. Cycle time is easy to measure, stand at a station with a stopwatch, or pull it from the machine, so it gets quoted. Lead time is harder because it spans the whole plant and most of it is invisible waiting nobody is timing. The path of least resistance is to report cycle time and hope it answers the lead-time question. It does not.
The other source of confusion is the third metric in the family: takt time. Cycle time is what the process does, takt time is what demand requires, and lead time is what the customer experiences. Keeping all three straight is worth a few minutes; the pairwise comparison of the first two is in our cycle time vs. takt time guide. Once the three are separated in your head, the improvement work sorts itself: takt sets the target, cycle time tells you whether stations hit it, and lead time tells you whether the customer is happy.
Getting both numbers right depends on measuring them from the same source instead of two different clipboards. When cycle times come straight from machine signals and lead time is reconstructed from live order and WIP data rather than from paper at week-end, the two stop contradicting each other and start explaining each other, you can see exactly how much of a nine-day lead time was queue versus work. That is the approach Harmony takes when it connects the floor into one operational layer with no rip-and-replace of the equipment; the plant in our CLS case study reads cycle time and lead time off the same live picture, which is what keeps the shop from quoting one when it means the other.