Multi-location phone systems: one plan for three or more sites

Why three sites is the tipping point

Two locations can limp along on two separate phone accounts with a forwarding rule between them. At three, the workarounds stop scaling. Staff keep paper lists of each other’s outside numbers, transfers between sites mean hanging up and redialling, the head office cannot see whether a branch is answering, and each site has its own after-hours message that nobody updates at the same time.

A single cloud system fixes this structurally rather than with rules. Every site is a group of extensions on the same platform; the network between them is the internet each site already has. The planning work is in deciding how the sites should appear to callers and to each other, and that is what the rest of this guide is about. It applies equally to a chain of clinics, a distributor with branches, a firm with regional offices, or a retailer with stores.

Decide how numbers are presented

There are three reasonable models, and most companies end up with a blend. The first is one published number for the whole business, with a menu or an AI receptionist that routes to the right site. The second is a local number per site, each ringing that site directly, with the head office number separate. The third is both: local numbers for people who know which site they want, and one main or toll-free number for everyone else.

The choice is about the caller. A patient calling a clinic wants their clinic, so a local number per site matters. A customer calling a distributor wants an order taken, and does not care which branch does it, so one number with routing by postal code or by the least-busy site serves them better. Write down who calls each site and let that choose the model.

Whatever the model, each site’s outbound caller ID should show the number that site wants called back on. A branch that shows the head-office number will have its callbacks routed through head office forever.

Build an extension plan that survives site four

Extensions across sites need a scheme, and it should be set before the first site goes live, because renumbering later is disruptive. The simplest is a leading digit per site — 1xx for head office, 2xx for the first branch, 3xx for the second — with a consistent pattern inside each range: x01 for the site’s main desk, x10 to x49 for staff, x50 up for rooms and shared devices.

Four-digit extensions cost nothing and give room for a fifth and sixth site without reorganizing. Reserve a range for people who move between sites and for roles that exist at every site, so that ‘the manager’ at any branch is always the same last two digits. Reception at every site being x00 is the kind of small consistency that stops people looking things up.

• One leading digit or block per site, assigned before any site goes live
• Consistent role positions inside each block (reception, manager, shared rooms)
• A separate block for roaming staff and head-office departments
• Spare capacity for at least two more sites than you currently have
• A written directory that is published to every desk phone and app automatically

Route by site, by time zone and by load

With sites on one system, routing becomes a company-wide decision rather than a per-site one. Calls to a site’s local number ring that site’s group, overflow to that site’s voicemail or, better, to another site that is open. Calls to the main number can be routed by the caller’s area code to the nearest site, by a menu choice, or to whichever site currently has the shortest wait.

Time zones matter across Canada. A company with sites in Halifax and Vancouver has four and a half hours of difference in opening times, which means the Halifax site can take early calls for the west and the Vancouver site late calls for the east. A schedule per site, and an overflow rule that looks at which sites are open, turns that into extended coverage without anyone working late.

After-hours rules should be defined once and applied per site with local variations — the same message structure, the same on-call mechanism, but the branch’s own hours and holiday list. A single person at head office maintaining the schedules avoids the branch whose holiday message still mentions last year.

Failover between sites

The advantage of multiple sites is that they back each other up, but only if the failover is configured. If a site loses internet or power, its desk phones go quiet; the system should notice and send that site’s calls elsewhere — to another site’s group, to staff mobile apps, or to a cloud voicemail with a transcript — without anyone having to log in and change anything.

Test this before you need it. Unplug a site’s router on a quiet afternoon and confirm that its local number rings where it should, that its outbound caller ID does not matter for the duration, and that staff mobile apps still register. Then plug it back in and confirm the site recovers on its own. A failover plan that has never been tested is a hope.

Emergency calling is per site and per device. Canadian VoIP providers must provide 9-1-1 service and warn customers of its limitations; the registered service address for each location is confirmed at setup, and the person responsible for keeping it current when a site moves or a device is shipped between sites should be named in the plan.

Roll out one site at a time

Do not cut over every site on one day. Start with the site that is easiest — good internet, a small team, a manager who is interested — and use it to prove the extension plan, the routing and the training. Its numbers port, its staff learn the apps, and the problems that surface are fixed before they are multiplied by the number of sites.

Then move to the next site, reusing the same configuration with local changes. Head office is often best done last, because it carries the main published number and the most complex routing, and because by then the process is familiar. Each site’s old service stays live until its port date, so nothing is lost in between.

Keep a single document listing every site, its numbers, its extension block, its hours, its failover targets and its 911 address. That document is what the next site is built from, what a new manager is handed, and what a specialist works from when something needs to change.