When the Lead Time Becomes a Range

How to plan, reorder and promise when your supplier's "three weeks" keeps moving.

The email from the supplier says three weeks.

Last time it said three weeks too. That order took five. The one before took two and a half. Same supplier, same product, same email. The goods simply arrived when they arrived.

So the buyer does what good buyers do. She adds a few days, quietly, just in case. Then a few more. And somewhere in a spreadsheet, a reorder date moves, a buffer grows, and a little more of the company's cash goes to sit on a shelf.

Here's the thing nobody says out loud. The problem isn't that the lead time is long. A long lead time you can plan around. The problem is that it has stopped being a number at all. It has become a range.

This piece is about planning with that range, honestly, instead of pretending it's still a number.

The short version. When supplier lead times become unpredictable, plan with the range, not the average. Measure how long each supplier and route really takes, including the bad deliveries. Use the typical time to decide when to reorder, and the spread to decide how much buffer to hold. Re-plan the same day an arrival date moves. And review the buffer every month, because safety stock that nobody reviews turns into dead stock.

Why does a lead time stop being a number?

Because every link in the journey has its own wobble, and the wobbles add up.

A supplier's production slot slips. A carrier misses a pick-up. A shipment sits in a customs queue. Peak season congests a port, or a warehouse, or a border crossing. A container is rolled to the next sailing. Any one of these is ordinary. Together, they turn "three weeks" into "somewhere between two and six".

Ocean freight shows it clearly, because it is measured so carefully. Sea-Intelligence tracks how many container ships arrive on schedule around the world. In May 2026 it reported 64.7%, the best month of the year. By August, it had fallen to 49.9%, the lowest since 2022, and the ships that were late averaged 6.81 days late.

Read that again. Half the world's container ships arrived off schedule, and the average late one was nearly a week behind. If your goods travel by sea at any point, some of that wobble is now sitting in your lead time, whether your supplier's email admits it or not.

And it isn't only ships. Road, air, customs and suppliers' own factories all have their own versions. The sea just measures it most publicly.

What does an unpredictable lead time actually cost?

More than a long one. That surprises people, so here's why.

Say you sell 120 units of a product a day.

If your supplier reliably takes 35 days, you reorder 35 days ahead and keep a modest buffer for demand. It's a long wait, but it's a known wait. You can plan your whole operation around it.

Now say the supplier usually takes 21 days, but sometimes 14 and sometimes 35. On average it's faster. But to avoid running out on the bad orders, you have to hold enough stock to cover those extra 14 days. At 120 units a day, that's roughly 1,680 units sitting in the warehouse, just in case. Paid for. Taking up racking. Waiting for a delay that might not come this time.

A late delivery you can see coming is a planning problem. A late delivery you can't is a cash problem.

There's a second cost, and it's sneakier. When lead times wobble, people protect themselves. The buyer pads the reorder date. The warehouse holds a few extra pallets "just in case". The sales team orders ahead for their best customer. Each decision is sensible. Together, they send a distorted signal up the chain.

Researchers have a name for this. In a well-known 1997 paper, Hau Lee, V. Padmanabhan and Seungjin Whang described the bullwhip effect: orders swing more than real sales do, and the swings get bigger the further up the supply chain you look. Your supplier sees your padded order, assumes demand has jumped, and plans accordingly. Their lead times stretch. Yours do too.

It can even become self-fulfilling. MIT's Stephen Graves points out that planned lead times are usually set by experience, often in reaction to the last extreme event, and padded rather than modelled. In a factory, a longer planned lead time releases work earlier, which crowds the floor and can make the real lead time longer still. Pad for the worst case, and the worst case gets a little more likely.

How do you plan with a range instead of a number?

You stop asking "how long does this supplier take?" and start asking "how long does this supplier take, and how much does that move?"

Here's a practical way to do it, without any new software.

1. Measure what actually happened. For every purchase order over the last six months, write down two dates: when you were told it would arrive, and when it really did. Do it per supplier and per route, because the same supplier can behave very differently by sea and by air. You're looking for the typical time and the bad tail, not just the average. An average hides exactly the deliveries that hurt.

2. Split the two jobs. The typical lead time tells you when to reorder. The spread tells you how much buffer to hold. Mixing them is how buffers quietly grow out of control, because every bad delivery gets baked into the reorder date as well as the stock level.

3. Buffer where it matters, not everywhere. A fast-selling product with an unreliable supplier needs a real buffer. A slow-moving product with a reliable supplier barely needs one. Spreading the same "two extra weeks" across every product ties up cash in places it does nothing.

4. Re-plan the day a date moves. When a supplier pushes an arrival back by five days, that's new information. The reorder plan, the transfers between your sites, the receiving schedule and the customer promises that depended on the old date all need to know today, not at next month's planning meeting.

5. Review the buffer every month. Safety stock is a decision, not a fixture. If a supplier's spread has narrowed, release the extra stock and get the cash back. If it has widened, raise it deliberately, not by accident.

6. Talk to your supplier in ranges too. Ask for their honest range, not their best case. Share your forecast so they aren't reacting to padded orders. The bullwhip runs both ways, and so does the fix: Lee and his colleagues recommended sharing demand and inventory information along the chain for exactly this reason.

None of this is complicated. It is just rarely done, because the numbers live in different places: promised dates in emails, real dates in the goods receiving log, and the buffer in someone's head.

What changes for the rest of the operation?

A range doesn't just change purchasing. It changes everyone downstream.

The warehouse stops getting steady arrivals and starts getting bunches: three late shipments landing in the same week as two on-time ones. Receiving needs to see what's coming, not just what's due.

The fleet has collections that move. Planning an inbound pick-up around an arrival date that slips twice is how trucks end up waiting at a gate.

And the sales team has to tell customers something. When the stock that's promised to them is somewhere on the road with a moving date, "when can I have it?" becomes the hardest question in the building.

That's why this is the first piece in a short series. The next two take those downstream problems one at a time: seeing all your stock, wherever it is and however it's travelling, and deciding what to promise a customer when you don't know when stock will land.

How this runs on one platform

Planning with a range needs three things in one place: what you have, what's already promised, and what's coming and when.

That's how stock is held in Depot WMS: what's on the shelves by warehouse and bin, which customer orders have already claimed it, and every open purchase order counted as inbound stock, with its destination and an expected receipt date taken from the order's delivery terms. A planner sees the whole position on one screen instead of rebuilding it from three.

Depot WMS also keeps both dates on every purchase order: when the goods were due, and when they were actually received. So each supplier's real lead time, and how much it moves, is on record, instead of being pieced together from old emails.

Replenishment works from that full picture. The reorder calculation takes in on-hand stock, stock committed to open orders, stock already on its way, sales velocity, vendor lead times and minimum order quantities, and proposes purchase orders into an approval queue. The system does the arithmetic. The buyer still makes the call, which matters most when the lead time is a range.

Goods that move in stages, say from a port to an import warehouse and then out to a regional site, are handled in Depot WMS as one chained movement. Each leg has its own status, and a delay on an early leg flows through to everything planned after it, so nobody finds out when a truck turns up to an empty bay.

And if a product does run short, Flow OMS doesn't have to turn the customer away. It can take the order as a backorder and release it for fulfilment as soon as the stock is received, so a late shipment needn't become a lost sale.

None of that makes a supplier more reliable. It makes the uncertainty visible, which is the only way to plan around it.

The short version

Lead times haven't just got longer. For a lot of businesses, they've stopped being predictable. Plan with the range: measure what really happens, reorder on the typical time, buffer for the spread, and re-plan the day a date moves.

An honest range beats a hopeful number every time. The hopeful number just takes longer to let you down.

Talk to us about planning around lead times you can't fully trust

Frequently asked questions

What is lead time variability?

Lead time variability is how much the time between placing an order and receiving it changes from one order to the next. A supplier that always takes 30 days has low variability. A supplier that takes anywhere from 20 to 45 days has high variability, even if its average looks reasonable. Variability matters more than length for inventory planning, because it is what forces a business to hold extra safety stock.

How does lead time variability affect safety stock?

Safety stock has to cover the deliveries that arrive later than usual, not the average one. The wider the spread of lead times, the more extra days of demand the buffer must cover. For a product selling 120 units a day, a lead time that can run 14 days longer than usual means holding up to 1,680 extra units to avoid a stockout on those orders.

How do you measure supplier lead time variability?

Compare the promised arrival date with the actual receipt date for every purchase order over a recent period, such as six months. Group the results by supplier and by route, then look at both the typical lead time and the slow tail, for example the worst one in ten deliveries. The typical time sets when to reorder, and the spread sets how much buffer to hold.

Why do businesses over-order when lead times are uncertain?

Each person in the chain protects themselves by padding: buyers order earlier, warehouses hold extra and sales teams order ahead. Researchers call the result the bullwhip effect, described by Lee, Padmanabhan and Whang in 1997, where orders swing more than real demand and the swings grow further up the supply chain. Sharing demand and inventory information with suppliers helps reduce it.

How reliable are container shipping schedules in 2026?

According to Sea-Intelligence, global container schedule reliability was 64.7% in May 2026 and fell to 49.9% in August 2026, the lowest since 2022. Vessels that arrived late in August were late by 6.81 days on average.

Should safety stock be reviewed regularly?

Yes. Safety stock should be reviewed at least monthly against the latest lead time spread for each supplier and route. When a supplier becomes more reliable, the extra buffer can be released to free up cash. When it becomes less reliable, the buffer should be raised deliberately rather than through ad hoc padding.

References

Sea-Intelligence (2026). Global schedule reliability for May 2026 the highest of the year. Global Liner Performance press release.

FreshPlaza (2026). Global container shipping schedule reliability falls to lowest level since 2022. Reporting Sea-Intelligence's Global Liner Performance data for August 2026, 30 September 2026.

Lee, H. L., Padmanabhan, V. and Whang, S. (1997). Information Distortion in a Supply Chain: The Bullwhip Effect. Management Science, 43(4), 546-558.

Graves, S. C. How to Think About Planned Lead Times. MIT Leaders for Global Operations Review.

More in this series

This is the first piece in Planning for the Unpredictable. Coming next: one stock picture, however it's travelling, and what to promise a customer when you don't know when stock lands.

For media inquiries, please contact:
contact@illuminate.ae