Thanks for asking, lead times are tricky. I’ll try to explain why.
For what it’s worth, we just shortened it to 60 days from 120 days, where it has been for quite some time. Throughout the fall we hit that issue exactly as you describe: a customer placed an order with some long lead time (between 90 or 120 days in the fall, I don’t have at hand the exact times we changed things), then we hit one of a series of internal delays that delayed their order past the lead time we predicted. Yet we didn’t change the lead time on the site. Why?
The lead time in the shop is an estimate on when we believe an order placed today will ship. For orders with a backlog, it’s an attempt to predict that ever-elusive shipping parity date. We review those numbers periodically and ask the question: “Would an order placed today ship within that time?” If based on all the information we have, we believe it would, the lead time stays the same. If we think it would ship significantly sooner, we shorten it, and if we have hit a delay that makes us think it should be longer, we extend it.
What happened with L5USA during the fall is not a single delay, but a series of different delays spread out over a number of months (component shipping delays among other things). So a particular delay would hit that would affect existing L5USA orders, let’s say it would delay them by a few weeks. At that point we’d look at the information we had on the nature of the delay, when it would be resolved, etc. and ask “Would an L5USA order placed today ship in 90 days? 120 days?” whatever the lead time was. Usually the delay wasn’t significant enough we felt it would affect new order lead time so we kept the lead time. At one point we did extend it to 120 days because we felt that was more accurate.
Fast forward a few weeks and then some new delay we didn’t anticipate would surface that would affect past orders. Again we’d review the delay, our info on when it would be resolved, and the current lead time and decide whether it was still accurate. For the most part we felt it was so we didn’t change it. I completely understand why L5USA customers were frustrated, we were frustrated too.
Fast forward to this week. We reviewed all the information we have, in particular in light of the fact that we had contacted all orders that required shipping updates (in case their address changed since they placed the order some time ago). The next batch of orders we would ship out after this batch would just ship out, they were recent enough that we didn’t need to ask for updated shipping from the customer. Based on that, combined with our L5USA stock, shipping throughput, number of orders, L5USA production schedule for the next month, etc. we felt confident enough to shorten the lead time from 120 days to 60. That means we feel confident that an order placed today would ship by July 16. Many internal signs are pointing to the actual lead time being shorter, but again, we are trying to pad things and be conservative, because there always seem to be new unforeseen shipping delays or other issues.
An “X days from today” style of lead time naturally shifts ahead each day. Could we pick an actual date instead (say Q3 2022 or July 2022)? We could and sometimes we have in the past, yet we found when we did that, especially when the date was more specific, we had to keep on top of it and update it much more often so it wasn’t inaccurate, as each month went by. We found it was more likely to be inaccurate, so we shifted to a sliding window.
Of course this approach is imperfect too, and it can lead to the kind of situation we hit this past fall. We don’t want to do that any more if we can help it, as it’s frustrating for us and it’s frustrating for you. This is why we are trying to move each product away from Just In Time fulfillment to a model where we hold a lot of stock. This is how we are ultimately going to resolve this issue and make things stabler and more predictable.