🎯 Quick Answer

What: Lead time is the total time from when a request is made to when it is done. In most processes, only a small part of that time is actual work. The rest is waiting. Much of that waiting comes from batches: points where work is held until there is enough of it to handle all at once.

Why it matters: Most efforts to reduce lead time focus on people: work faster, add staff, push harder. But if the time is spent waiting in batches between people, none of that helps. Every department can be fast while the process as a whole is slow.

How to apply: Follow real pieces of work and track two times: the time someone is actually working on it, and the total time from start to finish. Find every point where work waits for other work. Measure how long it waits there, ask why the batch exists, and shrink the biggest waits first. For example: meet more often, let a deputy sign routine approvals within set limits, and stop sending returned work to the back of the queue.

The payoff: Lead time drops, often by more than half, with the same people doing the same work.

At one plant, everyone agreed the engineering change process was slow. They just could not agree on who was slowing it down.

Some said drafting. Some said purchasing. Some said quality. It usually depended on which change they had been waiting on most recently. A change took about three weeks from the day it was requested to the day it reached the floor. Everyone agreed on the three weeks. Everyone also believed the delay was in someone else's department.

A shorter version of this story ran in September's field notes. It ended once the team found where the time was going. This version tells the whole story, including what the team changed next, because that second half is where the lead time came down.

Two clocks

Instead of arguing about who was slow, a small team followed twenty changes from start to finish. They timed each one with two clocks.

The first clock ran only while someone was actually working on the change: reading it, marking up the drawing, checking a part number, signing it. The second clock ran from the day the change was requested to the day it reached the floor.

The first clock rarely went past two hours. The second clock averaged sixteen working days.

That gap is the most useful number to have when you want to reduce lead time, and very few teams ever measure it. The method behind it, and the eight types of waste it sorts time into, is covered in The Waiting Is the Process. In this case, it did one thing first: it ended the argument. No one had expected the gap to be that big.

Where the sixteen days went

The time was not spent in any one department. It was spent between departments.

Each department's own work was quick. Every group in the process was fast, but the process was slow.

That afternoon, the argument about who was slow stopped. Nobody had been proven wrong; the question just no longer made sense. The cause was not a person or a department. It was that each change spent about fifteen and a half days waiting while busy people worked on other things. Until someone wrote that number next to the two hours of work, nobody had noticed it. The waiting just looked like the normal schedule.

Why each batch existed

Each of those four waits was a batch: a point where work was held until there was enough to handle at once. None of them was a bad idea when it started. Each one had a reason.

The weekly board existed because getting engineering, quality, production and purchasing into one room takes effort, so it made sense to do it once a week for everything. The signature waited because the manager had the authority and was only on site one day in five. The oldest-first queue was meant to be fair. The drawing sets existed because sending revisions out together was thought to save work for document control.

The team asked the same question about each one: what is this batch protecting, and is that worth the wait it adds to every change? In all four cases, the benefit was real but small. The cost, counted in days of waiting per change, was large.

What the team changed

No one was asked to work faster. No one was added.

The first clock did not change. The work still took under two hours. The second clock dropped from sixteen working days to about seven over the next two months.

Before claiming success, the team did two things worth copying. First, they compared the new lead times to the normal range of the old ones, to make sure the drop was not just ordinary ups and downs. Second, they checked what else had changed at the same time, such as staffing or a slower order book, that might explain the drop. Nothing else had changed. The smaller batches were the reason.

The same math at every scale

Every item in a batch waits for the whole batch. The first item in waits for the last. When a batch runs on a schedule, the average item waits about half the time between runs. So a weekly meeting means the average item waits half a week before anyone looks at it. Double the time between runs, and the wait roughly doubles. Cut it in half, and the wait roughly halves. The work itself stays the same.

The same math applies at every scale: a tray of signatures, a weekly review, orders released once a month, or a big push at the end of every quarter. When a plant is half idle in the first week of the quarter and working weekends in the last week, it is usually one batching problem showing up at both ends. The work arrives in big batches, so the floor gets too little work and then too much.

That is also why pushing people to work faster so often fails to reduce lead time. The work was never the slow part.

How to find the batches in your own process

In DMAIC, the step-by-step method at the heart of Lean Six Sigma, finding the waiting belongs to the Analyze phase, and shrinking the batches is often where the Improve phase starts. It is one of the cheapest ways to reduce lead time. It also tends to last, because no one has to keep working harder to keep it going.

I write one of these every week, based on what actually happens on projects rather than what the textbook says should happen. You can join the newsletter here.

FAQ

What is the difference between lead time and cycle time?

Lead time is the total time the customer waits, from request to delivery. Cycle time is used in more than one way. Many organizations use it to mean the time spent actively working on one item at one step. Others use it to mean the time between finished items. Because the definitions vary, agree on them in writing before you measure anything. The story above uses the simplest version: one clock for working time and one for total time.

Why doesn't working faster reduce lead time?

In most processes, the work is only a small part of the lead time. If a change takes two hours of work and sixteen days in total, making everyone twenty percent faster saves less than half an hour. The sixteen days barely change. The time is in the waiting, and waiting depends on how work is batched, queued and approved, not on how hard people work.

How does batch size affect lead time?

Every item in a batch waits for the batch to fill or for its scheduled turn. For a scheduled batch, like a weekly review, the average item waits about half the time between reviews. For a quantity batch, like a full cart or a full furnace load, the first item waits until the last one arrives. Bigger batches mean longer waits, and the waits add up across every batch in the process.

Are some batches necessary?

Yes. Some batches protect something real, usually a setup that takes a long time, such as heating a furnace, calibrating an instrument or getting a group together. The lean approach is not to get rid of those batches but to shorten the setup, because a cheaper setup makes a smaller batch affordable. If a batch protects quality, such as a review that catches errors, keep the check but cut the wait: run it more often, for less time each time.

How do you measure lead time without special software?

Two dates per item are enough: when it was requested and when it was done. Add a simple log of working time for a sample of items, even a tick sheet, and twenty items will show you the pattern. A spreadsheet handles the rest. The goal is not precision. The goal is to put the working time and the total time side by side, where the gap between them is obvious.