A reusable sleeve box programme that runs cleanly at one site gets approved for rollout and then degrades steadily as sites are added. The packaging does not change; the information does. Multi-site loops fail on counting and ownership far more often than on anything physical.
Define one standard, then allow local variation deliberately
Sites need a common definition of what an empty is, what state it is in, and what terminates a unit. Without that, an inter-site balance never reconciles. Once the definitions are shared, local layout differences are fine – the one page approach from the loop procedure scales if the definitions are held centrally and the layout detail is local.
Transfers are where the numbers die
Units moving between sites are nobody’s issue and nobody’s receipt. At one site that does not matter; at five it is the whole discrepancy. Put inter-site transfers on their own line in the weekly count from day one, even when the volume is small.
Name an owner per site, and one overall
Each site needs a person who sees the count weekly. The programme needs one person who sees all five and is allowed to escalate. Without the second role, discrepancies are resolved locally and optimistically, and the central picture is fiction.
Expect uneven loss and investigate the outliers
National loss averages hide everything. One site will be losing units at several times the rate of the others, usually for a mundane reason – a yard, a shift, a customer, a door. Rank sites monthly and ask about the top one; that single habit recovers more units than any system.
Plan the wash capacity site by site
A programme sized on total volume ignores that washing capacity is local. The site that peaks hardest will be short of drying capacity first, and it will release units damp – which is where hygiene problems start. Size cleaning per site against that site’s peak, not against fleet average.
Spare stock needs somewhere to live
Multi-site programmes need a buffer to cover the site that runs short. Decide where it sits and who can call on it. Without an agreed buffer, sites hoard, and hoarding converts a small shortage into a system-wide one.
Roll out in a sequence that tests the weak link
Add the easiest site first if you need momentum; add the hardest first if you want to find out whether the design holds. Both are defensible – but decide deliberately rather than rolling out in order of whoever asked first.
What tends to work
Two habits make multi-site programmes work, and neither is about packaging. First, a monthly ranking of sites by loss rate, with the worst one asked about rather than reported on – that conversation recovers more units than any system. Second, a quarterly call where the site owners compare numbers directly, which spreads the good practices faster than any central document.
Programmes that fail at scale usually did neither. The packaging was fine, the reporting was centralised rather than shared, and local improvisation filled the gap.
What we can do
We supply returnable sleeve containers for multi-site programmes and can supply split deliveries and marked batches so each site’s stock is identifiable. Tell us how many sites and who will own the count, and we will flag what usually goes wrong at that scale.