Lead-Time Variability: What a Resilient Supply Chain Does Differently
After eighteen months of pandemic-era disruption, most planners have stopped pretending lead times are a fixed number. A supplier who quoted four weeks now delivers in three, then nine, then six. The mistake is not that lead times vary — they always have. The mistake is designing a supply chain as if they did not. Below is a side-by-side look at how two otherwise similar operations handle the same uncertainty, and why one keeps its shelves stocked while the other keeps apologizing to customers.
What bad looks like
The struggling organization treats the average lead time as the truth. If shipments arrive in four weeks on average, it plans to a four-week horizon, sets reorder points off that single figure, and expresses surprise every time a delivery is late — which, by definition of an average, is roughly half the time.
Safety stock is a flat percentage applied across every item, unrelated to how erratic each supplier actually is.
When a shipment slips, the response is an expedited air freight order that quietly destroys the margin on that product line.
Demand and supply data live in separate spreadsheets owned by separate people who meet only when something has already gone wrong.
Nobody measures lead-time variability, so nobody can tell whether a supplier is genuinely improving or just got lucky last quarter.
What good looks like
The resilient organization treats lead time as a distribution, not a point. It knows the average and, more importantly, the spread. It sizes its buffers against that spread and concentrates protection where variability actually lives, rather than smearing inventory evenly across the catalogue.
Measure the variability, not just the mean. Track the standard deviation of lead time per supplier and per item. An item that arrives in 28 days plus or minus 2 needs very different protection from one that arrives in 28 days plus or minus 14.
Set safety stock against demand-during-lead-time. Buffer for the combined uncertainty of how long supply takes and how much you will sell while you wait — not for a flat rule of thumb.
Segment your suppliers. Protect the volatile, high-impact lines heavily; carry less on the steady, low-impact ones. Uniform buffers waste cash and still leave you exposed.
Build dual or regional sourcing where it pays. A second qualified supplier — ideally on a different transport lane — turns a single point of failure into a manageable risk.
Share signals upstream. Give suppliers a forward view of demand so they can pre-position capacity instead of reacting to a purchase order that lands as a surprise.
The shift in mindset
The real difference is not a fancier system. It is accepting that variability is a permanent feature of the environment and designing for it on purpose. The resilient operation spends a little more on buffers and second sources, and in return it stops spending a great deal on expedited freight, lost sales, and firefighting. During the recovery, the hybrid teams that fared best were the ones that made these trade-offs explicit — written down, costed, and reviewed — rather than leaving them to whoever happened to be on the phone with the carrier that morning.
You cannot eliminate lead-time variability. You can decide, in advance and with clear eyes, exactly how much of it you are willing to absorb and where.
When you are ready to put real numbers behind these decisions and harden your sourcing, XNM's procurement, sourcing & contract management advisory can help you segment suppliers, size buffers, and qualify second sources.