The short answer
- Orders, artwork and customers must be shared; production scheduling can be local
- Two sites with separate order lists is two businesses, not one
- Stock has to be tracked by location or you will promise what is in the other building
- Decide job routing by a stated rule, not by whoever answers the phone
- Communication, not software, is what fails first
A second location arrives before anyone plans for it. A shop takes on more space, opens a counter across town, or has a designer working from home, and the systems that worked for one building start producing small daily frictions that nobody quite attributes to the change.
The pattern is consistent: the work distributes fine and the information does not.
What actually changes with a second site?
Very little about production and a great deal about visibility. Each site can print perfectly well on its own. What stops working is anyone knowing what the other one is doing.
| Question | One site | Two sites |
|---|---|---|
| Where is this job? | Ask across the room | Phone the other building |
| Do we have those blanks? | Look | Look, then phone |
| Who is quoting this customer? | Obvious | Two quotes, different prices |
| Is the press free Thursday? | Look at the board | Which board? |
| What did we charge last time? | Ask the person | The person is elsewhere |
Every row is a phone call that did not exist before, and every phone call interrupts two people. That is the real cost of a second site, and it is why the answer is shared information rather than more coordination.
Two sites do not need two systems. They need one system and two boards.
One order list
The non-negotiable. Every order, from either site, in one place, visible to both. The moment each location keeps its own list you have two businesses sharing a name — and customers discover it, usually by being quoted twice at different prices.
That includes customer records and artwork. A customer who ordered from one site and calls the other should be recognised, with their history and their logo available. Anything less makes the second site feel like a different company to the person buying.
Holding orders, customers and artwork centrally in your back-office system is what makes that work. What can stay local is the day-to-day production board — how each site sequences its own presses is its own business.
Stock in two places
Stock is where multi-site most often goes wrong, because a single total is worse than useless. A system reporting forty navy larges when twenty-five are in the other building will cheerfully let someone promise a delivery they cannot make.
Stock has to show
- Quantity per location, not just a total
- What is committed to a job versus genuinely free
- Anything in transit between sites
- Which site an incoming supplier delivery is going to
Transfers between sites should be recorded as movements rather than adjustments, or stock drifts. Two shops in the same town moving garments back and forth informally will be materially wrong within a month, and the error only surfaces when a job is short.
The simpler answer, where it works, is to not split stock at all — one site holds blanks and ships to the other. That trades a little transit time for removing an entire category of problem, and for most two-site shops it is the better arrangement.
Routing a job
Decide by a stated rule rather than by whoever took the order, because otherwise jobs land where they were sold rather than where they should be made.
| Route by | Rule |
|---|---|
| Capability | Only one site has the process — no decision to make |
| Capacity | Whichever has press time before the due date |
| Proximity | Nearest to the delivery address |
| Stock | Where the blanks already are |
| Specialism | Where the people who do this work best are |
Write the order of those tests down. Without it, routing is decided by habit and the result is one site quietly overloaded while the other has gaps — which shows up as a falling on-time rate at one location and nobody able to say why.
People and access
Everyone needs to see everything and change only what their role covers. That is more important across sites than within one, because the informal check — someone noticing a colleague about to do something odd — does not happen at a distance.
Individual logins are the foundation. When a price changes or a date moves, somebody at the other site needs to be able to see who did it and when, and that is impossible with a shared account. The four roles transfer directly; what changes is that the audit trail matters more.
The same applies to a home-based designer or a part-time bookkeeper. Give them a scoped account rather than emailing files back and forth, so the work stays attached to the job and access can be ended cleanly.
What breaks first
Not the software. Three human things, in roughly this order.
Nobody is accountable for the whole. Each site optimises itself, and jobs that need both fall between. Name one person responsible for the flow across sites, not just within each.
The two sites stop talking. Weekly contact between the people running each is worth more than any dashboard. A short standing call covering what is coming, what is stuck and what is spare prevents most of the coordination failures.
Standards drift. Two sites decorating the same customer's garments will diverge unless the specification is shared — which means the same recorded print size, placement, ink references and colour mix against the customer, not in one person's notes. A reorder produced at the other site should be indistinguishable from the first.
Before opening a second site
- One order list, one customer list, one artwork store
- Stock reported by location
- A written routing rule
- Individual logins with an audit trail
- A named owner for cross-site flow
- A weekly call that happens whether or not there is a problem
Do those six and a second site is capacity. Skip them and it is two shops arguing about whose customer this is. The rest of the software archive assumes both sites are looking at the same information.



