Six weeks to peak: warehouse peak season preparation

By mid-October, warehouse peak season preparation mostly comes down to sorting the ‘before peak’ list into three piles. Starting with the whiteboard.

It’s almost the middle of October, and you know just as well as I do that there’s a whiteboard in the office with a ‘sort out before peak’ list on it that’s been growing steadily since August. Some of it is still perfectly doable. Some of it is a lovely idea for February (but not before then). And some of it, if you try it now, will make such an impression that everybody will still be talking about it at next year’s Christmas party. And, just to be clear, not in a good way.

The trick at this point is sorting one thing from the other – and being fairly ruthless while you’re doing it.

So, for your fun and convenience (you’re welcome), I’ve split your peak season checklist up into three piles. The first is stuff you can still fix, safely, and before peak starts to hit. The second is stuff you should absolutely fix, just not right now. And the third is stuff that you should leave well alone, however tempting it might look to mess with on a quiet Tuesday afternoon.

Even if you’re not running Dispatcher WMS, most of this warehouse peak season preparation will apply to you, too. The screens might look a bit different, but the logic will probably not.

Pile one: fix it now

The test for anything in this pile is threefold and quite simple enough: Can you change it in a day or two? Can you see straight away if it’s worked or not? And can you quickly put it back the way it was if it hasn’t? If the answer to all of those three questions is a yes, crack on.

Replenishment is usually the best place to start, because it’s where peak tends to start stamping on things before it hits anything else. Your min/max levels and replen triggers were probably set up for a normal week, and quite possibly they were set up for last year’s normal week, at that. Have a look at last November’s top movers (or your predicted top movers, if you have them) and ask yourself if the pick faces they’re pointed at can actually hold enough stock to get through a shift at peak volumes. If they can’t, your pickers are going to spend a good chunk of Black Friday standing at empty locations waiting for someone to bring them more of what they need. And nobody is going to enjoy that, much, are they? Nudging pick face trigger points up a bit on your fastest lines is a small, easily reversible change that can make a surprisingly big difference when and where it matters the most.

Then have a look at your putaway rules. Most WMS will send stock exactly where the rules tell it to, which is SUPER. Right up until the rules it’s using are for a warehouse that hasn’t looked anything like that in real life since about last spring. If a couple of big new clients or product lines have arrived since anyone last touched the aforementioned putaway rules, check that fast-moving stuff is going to be directed somewhere sensible, and not three aisles back, behind the Christmas decorations, and underneath the mezzanine with that dodgy light (get that fixed too and any others in the warehouse). Tweaking a handful of rules for lines that you know are going to move quickly is absolutely fine. Rebuilding your whole putaway strategy is something else entirely (and it’s in pile three, so put it down).

While you’re in sort-out mode, get your high-velocity locations counted. Going into peak with stock figures that you trust saves an awful lot of “but the system says there are twelve” conversations during December.

Next up, new people. If you’re bringing in agency staff, make sure that their user accounts, profiles and handheld logins are all set up and tested now. Rather than at six am on their first morning, while you’ve got forty new starters queuing outside the office door. Check that there are, in fact, enough handhelds. Check that they all actually charge and work properly, and while you’re at it, make sure that somebody knows where the spare batteries went last year. Do you know you can actually ‘rent’ additional kit from some kit providers? But do it early if you need to, in case everyone else needs it too.

Carriers are another big one. Peak can be exactly when carriers choose to change their collection times, add services, move cut-offs and occasionally update their label formats with very little warning. Test your label printing and carrier integrations end to end, with real orders, before volume starts to increase. If a label format has changed and nobody’s noticed, you’d much rather find out on a Wednesday afternoon in October than when you’ve got three trailers waiting to be dealt with outside.

And finally, a bit of housekeeping. Plenty of systems are carrying years of old order history, completed tasks and audit data that nobody’s looked at since it was created. Archiving what you no longer need can take a fair bit of weight off the database – before you ask it to work flat out for six weeks. Do it with your IT team (or ours), do it now, and then check everything afterwards.

Pile two: fix it, but not until peak is over

This is probably the hardest pile to leave alone, because everything in it is a perfectly wonderful idea.

Things like re-slotting the whole warehouse around this year’s velocity, reworking your wave or release strategy, rethinking the returns process that’s been bothering everyone since spring, or finally getting round to sorting out all of the workarounds that your team have built up over the past year. All these things are very sensible, all of them are very noble, and all of them are absolutely worth doing. But you really don’t want to be halfway through any of them when things start to get busy.

What you can do now, though, is write everything down. Peak is brilliant at showing you exactly where the system and the operation have drifted apart, because it pushes everything (and usually everyone) to the edge, all at the same time. Keep a proper list as you go, with dates, rough volumes and what actually happened (compared to what you were expecting/hoping would happen), and you’ll walk into January with the best evidence that you’ll have all year for what needs changing.

That list is more or less the starting point for a configuration review, and it’s far more useful than your management team trying to remember in February what it was that went hideously wrong on December 4th.

Pile three: just… please don’t

Some things look small from the outside, but on closer inspection, they turn out to be anything but.

Upgrades are at the top of that list. Even a minor version change touches on more than you’d expect, and peak is not the time to find out which of your customisations relied on the old behaviour. In order to be absolutely sure that everything works post a change like this, you’ll need to test absolutely everything, and that takes up a lot of time. So, unless there’s a real security reason, or your vendor is telling you very firmly otherwise, it can wait.

New integrations are next. A new marketplace, a new carrier, a new client who’d like to go live “before Christmas if possible”. It might well be totally possible. Whether it’s smart is a separate conversation altogether, and it’s one that’s worth having while everybody who will be affected by such a decision is in the same room, at the same time. And that’s because we all know that the people who’ll be dealing with it at two in the morning on a cold, wet Saturday the week before Christmas are rarely the same people who agreed to it in the first place.

Then, there’s automation. If the new sortation or robots aren’t already live and bedded in by now, the bit of peak that you wanted them for has probably gone. Bringing new kit online in November tends to mean running two processes side by side at exactly the time you can least afford to do so.

Drawing a line under your warehouse peak season preparation

Most operations that come through peak in decent shape have one thing in common in their warehouse peak season preparation: they agreed a date after which nothing would be changed. The exact date matters less than the fact that everybody knows when it is, including IT, including the account managers, and including whichever client is about to ask for “one small thing”. Mid-November is a fairly typical choice. After that, anything new goes on the January list, and the only changes that get made are the ones that fix something that’s broken enough to stop proceedings.

It’s not the most exciting rule in the world, but as with so much of what we talk about, being a bit boring is probably the smartest and most effective thing that you can do to keep things running when push comes to shove.

If you’d like a bit more on the longer run-up, our peak season prep article – Christmas in July from the summer covers the stuff that needed doing before now. It might be worth bookmarking for next year?

A second pair of eyes on your warehouse peak season preparation

At Socius24 Limited, we’d much rather help you find and fix problems in October than help you put out a fire in December. So, if you’re looking at your own peak list and aren’t quite sure which pile something belongs in, we’re always happy to have a no-obligation chat.

You can get in touch with us here.