What it means
Lead time is the time from a stated start to a stated finish: a purchasing team may start its clock when it releases a purchase order and stop when goods become available in stock, while another team may measure shipment to dock arrival. Lead time variability asks how much those measured times move around.
If a supplier repeatedly delivers in 10 days its lead time is stable, whereas deliveries of 5, 10 and 20 days are much less predictable even if the average appears workable. The distinction matters when planners set reorder points, because they need enough stock to cover demand while replenishment is in progress, and an occasional long delay can cause a stockout if the plan assumes only average time.
Safety stock can protect against uncertain replenishment, but it ties up cash and storage. ASCM describes lead time variability as an input to safety-stock calculation, and the amount depends on service target, demand and modelling assumptions.
Measure comparable orders rather than pooling everything without thought, since express freight and standard freight have different time patterns and combining them may produce a number that does not describe either service. Start and stop points should be stable, because a team that switches from dock arrival to shelf availability changes the measured lead time, and the apparent variation may be a process-definition problem rather than supplier performance.
Record order date, promised date, actual receipt and ready-for-use date where relevant, as the promise date helps assess reliability against a commitment while the actual order-to-ready time measures what planners experienced. A common summary is the standard deviation of observed lead times, which reports spread in the same unit as the lead time, such as days, while a histogram or percentiles show the shape and unusually long tails.
For example, three fictional orders take 8, 10 and 12 days, so their mean is 10 days and the population standard deviation is approximately 1.63 days, and a different set with a wider range would have a higher spread. Lead time variance is a related statistical term, often the average squared deviation from the mean, with a unit of squared days unlike standard deviation's days, and because some businesses also use "variance" informally for a difference from plan, the metric should be defined.
Seasonality may change the distribution, so if holiday shipments routinely take longer, maintain separate plans or forecasts for that period rather than treating a predictable shift as random noise. Separate variability from a biased promise, since a supplier may always take 12 days while promising 7, which makes delivery time consistent but promise accuracy poor.
The impact depends on demand during the wait, as uncertainty in a rarely sold item may be tolerable while the same delay for a fast seller may empty shelves, and because demand variation can interact with lead time variation, ASCM's discussion of combined safety-stock calculations, which notes that assumptions about statistical independence matter, is a reminder not to add buffers or use one formula blindly. A planner might respond by holding stock, ordering earlier, asking for better visibility or qualifying another supplier, and each response carries a cost, so compare service improvement with working-capital and logistics cost.
Supplier conversations should focus on process stages, because a precise timeline of time lost before dispatch, in transport or during receiving identifies a fix more clearly than a single late-order percentage. Track performance over a consistent window showing the mean, standard deviation, sample size and percentage of orders beyond a practical threshold, mark major changes in supplier, route or product mix, and treat lead time variability as a way of turning uncertainty into a measurable planning input.
In practice
Real-world examples.
Example
A supplier usually delivers in 10 days but occasionally takes 18, making a fixed reorder schedule risky. The buyer sets the reorder point using the longer observed delays, not the average alone.
Example
A buyer separates standard and express orders before measuring lead time spread. Each service then gets its own reorder plan.
Example
A warehouse measures arrival-to-shelf delays to see whether receiving work adds uncertainty. It finds the delay sits inside the building, not with the supplier.
Formula
Calculation
Population standard deviation of lead time = square root of [sum of (each observed lead time - mean lead time) squared / number of comparable observations]. Use an appropriate sample formula for sampled data.
Worked example. Supplier A delivers three comparable orders in 8, 10 and 12 days.
- Mean = (8 + 10 + 12) / 3 = 10 days.
- Deviations from the mean are -2, 0 and 2; squared they are 4, 0 and 4, which sum to 8.
- Variance = 8 / 3 = about 2.67 days squared, and standard deviation = square root of 2.67 = about 1.63 days.
Supplier B delivers in 6, 10 and 14 days, so it has the same 10-day mean.
- Deviations are -4, 0 and 4; squared they are 16, 0 and 16, which sum to 32.
- Variance = 32 / 3 = about 10.67 days squared, and standard deviation = about 3.27 days.
Both suppliers average 10 days, but Supplier B needs a larger safety buffer because its deliveries are about twice as spread out.Case study
Seen in the real world.
In this entirely fictional case, Alder Parts records the full order-to-ready time for twenty routine purchases. It finds a few unusually long waits despite an acceptable average. The team checks whether delays arise in transport or internal receiving before changing stock levels.
It documents the sample and the clock definition rather than attributing every delay to the supplier. The buyer then shares the timeline with the supplier and agrees to track the same stages monthly. In this invented story the long waits turn out to cluster in customs clearance for one route, so the team adds a second route for urgent items rather than raising stock across the board.
Watch out
Common mistakes.
- Confusing an unpredictable lead time with a merely long average.
- Mixing different shipping services in one unlabelled sample.
- Treating a safety-stock equation as reliable without checking assumptions.
Questions
People also ask.
How is it measured?
Collect comparable elapsed times and show their distribution, often with standard deviation and percentiles.
Is it the same as being late?
No. A consistently slow supplier may have little variation but still miss a promised date.
Why does it matter?
Uncertain replenishment can cause stockouts or require extra inventory and contingency plans.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related