Why the start date changes the whole production plan
An event order works backwards from a date that many people know about. An onboarding order works backwards from a date that one specific person knows about, is anxious about, and will mention more than once. The practical consequence is that the buffer which normally absorbs a delay — shipping something early and letting it wait — is largely unavailable. A parcel sitting in a reception cupboard for nine days does not read as preparedness; it reads as an administrative error, and the new starter discovers the company’s internal processes instead of meeting the team.
That constraint also inverts the usual order of operations. In most merchandise projects the artwork is finalised late because the quantity and the budget are settled early. Here the sequence has to be different, because the sizes arrive late, the sizes affect the specification, and the specification affects the lead time. A programme that asks for sizes after the order is placed has accepted a schedule it cannot hit. The workable order is: fix the design and the garment type first, hold a stock of core sizes, then place a rolling series of small replenishment orders against confirmed start dates as they arrive.
Three ways the rolling programme is usually organised
The first is a fixed starter order placed once, sized by historical mix, held centrally and sent out as people join. It is the cheapest per unit because setup is paid once and every later order is a small repeat. It fails when the mix is wrong, and the mix is always wrong at the start of a new programme because there is no history to size it from.
The second is a per-hire order placed against each individual start date. It gives an exact fit and an exact arrival, and it pays setup on every order unless the design is unchanged, which makes the unit cost noticeably higher and the schedule much tighter. It works when start dates are spaced far enough apart that each order can clear the full production window.
The third is a hybrid: a base stock of the common sizes in the common roles, plus a small rolling order for everything else. This is the structure most programmes arrive at after a year, and it is the one that survives growth, because it converts a fixed-date problem into a mixed problem with a fast lane and a slow lane.