What breaks when volume doubles
500 parcels a day is a personality. 5,000 is a different job. Volume does not create new problems. It removes the hiding places for the old ones.
500 parcels a day is a personality. Someone knows every ugly SKU. The 3PL account manager still picks up. Exceptions fit in a group chat. A missed scan is a story, not a statistic. The operation feels like it is under control because a few people are holding it.
Then you have a good quarter. Or a marketplace feature. Or a second country. 5,000 is not ten times 500. It is a different job. The people who held it cannot hold it. The group chat becomes noise. The personality does not scale. Only a process does, and you may not have one. You had heroes.
I've watched this from both sides of the dock. The volume looks like a win in the Monday meeting. By Thursday the win is a queue. By the next month the win is a reputation problem.
What breaks when volume doubles is not the warehouse. It's every place you were using a person as a process.
The things that looked fine
Cut-off. At 500 you can slip. A late order still makes the van. At 5,000 the van is full, the airline build is closed, and "just this one" is how you miss a connection and pretend it was the carrier.
Inventory accuracy. A little lie in the bin is cute when you can walk over and look. At volume, the lie becomes oversell, then marketplace penalties, then a finance meeting about why revenue isn't cash.
Address quality. You could hero a bad address when it was twelve a day. You cannot hero 120. Last mile will fail them in a batch, and your customer service team will discover, in public, that you never had a failed-delivery loop. You had a nice person with a phone.
Customs files. One broker, one style of invoice, one person who "just knows" which description a particular officer likes. Volume turns that folklore into a pile of holds. Holds do not care that you were busy being successful.
Exceptions stop being stories
This is the one that surprises people. Your exception rate might not even get worse. It just gets numerous. 8 percent of 500 is a morning. 8 percent of 5,000 is a department you forgot to hire.
If exceptions live in a hero, the hero drowns. If they live in a WhatsApp group, the group becomes a place where problems go to be acknowledged, not solved. If they live in a 3PL ticket pile with no SLA you actually read, you will learn about them from customers.
I've been the person on the 3PL side adding night staff because a brand doubled and briefed nothing new. Labour can be added. A decision tree cannot be added at 11pm. The partner will apply the tree they have, which is the tree you wrote when you were small, which is not a tree. It is a vibe.
Partners and software lag the volume
A 3PL will take the extra cartons. That is their commercial instinct. Whether they can run your exceptions at the new size is a different question, and they may not know the answer until you are already in it. Ask them, before the spike, what breaks in their process at 2x. If they only talk about floor space, they are not thinking about your 8 percent.
Software is worse. A WMS configured for a quiet floor will not reveal its stupidity until the scan queue is the bottleneck. Workarounds that were "temporary" at 500 become load-bearing walls at 5,000. You will not have time to rebuild them during the spike. You will have time to hate them.
Volume does not create new problems. It removes the hiding places for the old ones.
What to fix before you wish for growth
Write the exceptions. Owner, time limit, default. Do it while the pile is still small enough to see. That is the whole trick. The work is cheaper now than it will be when it is famous.
Define cut-off as physics, not optimism. Define "despatched" as a real event. Define failed delivery as a loop, not a mood. Define who can change a live order. If those sentences only exist in one head, volume will find that head and take it offline.
Staff the ugly, not just the belt. The belt is easy to hire for. The decisions are not. If your plan for doubling is "more pickers," you have a plan for the 92 percent. Plan the 8 percent or it will plan you.
And stop using last month's green SLA as proof you are ready. Last month you were small. Small is a different sport.
The goal was never more orders
The goal was an operation that still makes the same decision at 5,000 that it made at 500, without needing the same three people to be awake.
Want the volume. Just don't confuse a good quarter with a process. Double on paper first. Walk the handoffs. Name what will snap. Then go get the orders. The orders will come. The hiding places will not.