TEAMS PHONE MIGRATION

PBX out, Teams in: the part of the migration that breaks most projects

By Radboud Heinink | CyberTwice | Reading time: 8 minutes

A friend of mine renovated his kitchen last year. He spent six months planning. Picked the cabinets, agonised over the countertop, ordered an Italian range that cost more than my first car. The contractor showed up on day one, ripped out the old kitchen, installed everything beautifully. On day three he tried to cook for his family. The new range wouldn’t ignite.

Nobody had checked the gas line. The old kitchen ran on a small-bore pipe that fed a basic four-burner stove. The new range needed double the capacity. Six weeks of waiting for the utility company, a permit, a wall opening, and roughly fifteen thousand euros later, the kitchen finally worked.

This is exactly what happens in the majority of Teams Phone migrations.

The whole industry talks about Teams Phone migration like it’s a phone number problem. RFPs focus on porting, SBC sizing, direct routing topology, calling plans, emergency services compliance. Discovery sessions cover concurrent call volume, codec preferences, queue overflow rules. The plan looks complete. Then somewhere around week three of the cutover, someone walks up to the warehouse door, presses the buzzer, and nothing happens. Or the lift emergency phone fails its monthly test. Or the security desk realises they can’t see the camera feed when the front gate intercom rings.

The phone numbers were the easy part. The peripheral devices, the ones nobody scoped, are the gas line.

What gets planned, every time

Ask any Teams Phone migration team what’s on their list and you’ll hear the same things. Number porting strategy. Direct routing or operator connect decision. SBC vendor selection. Microsoft Calling Plans budget. Auto-attendants and call queues. Teams device strategy for handsets, headsets, and conference room kits. Compliance recording, if you’re regulated. E911 and emergency address mapping. A parallel run window if the team is being careful. Cutover weekend logistics.

This is the visible kitchen. Cabinets, countertops, appliances. All of it gets attention because it lives in the phone system mental model. It’s what the PBX vendor sold you. It’s what the Teams partner is comfortable scoping.

What never makes the list

Walk through any building that’s been running for more than five years and you’ll find a quiet ecosystem of communication devices that have nothing to do with the phone on a desk. Door intercoms. Video door stations at reception. Paging systems with overhead speakers across a warehouse or production floor. Lift emergency phones, which by the way are legally required in most jurisdictions to both call out and accept inbound calls. Loading dock buzzers. Gate intercoms in the parking garage. Cleanroom paging. Healthcare nurse-call systems. Conference room ceiling speakers. Production floor PA for shift changes. Forklift hands-free comms in larger distribution centres. Visitor management kiosks with a “press to call reception” button.

That’s a dozen device categories before you even get specific to the building. Each of them is a phone. Each of them was previously connected to the PBX. None of them is on the migration project plan, because none of them is owned by IT.

That’s the gas line.

Why this keeps happening

Three reasons, mostly.

First, these devices were installed at different times, by different vendors, often years apart. The intercoms came in with a 2014 access control project. The paging system was added in 2018 when the warehouse expanded. The lift phones are part of a building services contract IT has never seen. The PBX absorbed all of them silently because that’s what PBXs did. Nobody at the migration table thinks of them as phones.

Second, ownership is split. IT owns the phone system. Facilities owns the doors and the paging. Building services owns the lifts. Security owns the cameras and the gate. In a typical company those four groups don’t sit in the same meeting until something breaks. Migration discovery rarely includes a walk through the building with the head of facilities, which is exactly the conversation that would surface every one of these devices in twenty minutes.

Third, the devices are old, reliable, and invisible. They’ve been doing their job for ten or fifteen years. Nobody touches them, nobody documents them, nobody puts them in the asset register. Out of sight, out of plan.

“IT owns the phone system. Facilities owns the doors. Building services owns the lifts. Security owns the gate. None of them sit in the same meeting until something breaks.”

The day the bill arrives

Here’s what missing the gas line looks like in real life.

The lift emergency phone goes silent on cutover Saturday. The building manager doesn’t notice until Tuesday’s monthly compliance test. The lift now fails inspection. Best case, you have a week to get it sorted. Worst case, the lift gets taken out of service in a building with reduced-mobility tenants. That’s a regulated, public failure.

The reception video intercom stops working. Visitors stand outside in the rain. Couriers can’t deliver. Someone improvises by taping a phone number to the door and asking people to call. It looks shabby for a month while procurement tries to find a replacement that fits the existing wall mount.

The warehouse paging fails during a near-miss safety incident. The shift supervisor can’t reach the floor. The incident report flags the comms failure. Insurance asks questions. Health and safety opens an investigation.

Then comes the cost. Replacing all of these devices just to make them speak the same protocol as your new system runs anywhere from five to ten times the cost of integrating what you already have. A new SIP video door station costs roughly €1,200 installed. You’ve got six of them. That’s €7,200 just for the doors. Multiply across paging, lifts, and outdoor intercoms and you’re looking at fifty to a hundred thousand euros in unplanned capex, plus weeks of project delay, plus the operational pain while it all gets coordinated.

All of that, because the gas line wasn’t on the plan.

The “we’ll just put a SIP gateway in front” trap

This is the moment someone in the room says “we’ll front-end it with a basic SIP gateway and route the calls back into Teams.” It sounds reasonable. It’s also an expensive shortcut.

Basic SIP gateways are audio-only. They terminate the SIP leg, hand audio into the platform, and that’s it. No video. No camera feed. No two-way visual confirmation when someone rings the bell. No paging zone management. No identity flow. You’ve solved the technical problem of “the call connects” while introducing a worse user experience than what the company had before.

A 2014 video door station with a colour camera and the ability to show the visitor on a guard’s screen now becomes a beep on someone’s headset. The receptionist used to see who was at the door before deciding to let them in. Now they can’t. Multiply that across every intercom in the building and you’ve quietly downgraded your physical security posture as a side effect of the phone migration.

Same story with paging. Basic gateways treat overhead speakers as a single audio destination. Real paging needs zones, priorities, scheduled chimes, emergency override, and the ability to page from a Teams client to the right speaker group without the user having to think about which one. Not a nice-to-have. A core requirement in any building over a certain size.

Basic SIP gateways are the equivalent of running the new Italian range on the old gas line. Technically there’s a flame. Practically, you’ve capped what the kitchen can do.

What good looks like

The right version of this is unglamorous. You walk the building before the migration plan is signed off. You list every wall-mounted device that makes a noise or accepts a call. You ask facilities, security, and building services to do the same. You map each device to a known integration path inside Teams: the door rings a Teams team with the camera feed visible alongside the call, the paging happens from a Teams client to the correct zone, the lift phone calls out and accepts inbound calls without anyone replacing the lift hardware.

No rip-and-replace. No parallel system that “we’ll consolidate in phase two.” No quiet downgrade of the physical security stack to make the phone project look on-budget.

That’s the whole point. The migration plan should match the building, not the other way around.

Where to start

Before the next vendor meeting, do the walk. Take an hour. Bring someone from facilities. Note every device. Photograph the labels. You’ll find more than you expected, and the list itself will reframe the conversation with whichever partner is doing the migration.

We put together a checklist that covers the twelve categories of peripheral devices that nearly always get missed in Teams Phone migrations, with the questions to ask for each one before the project plan gets locked. It’s the document we wish we’d had on every migration we’ve walked into halfway through.

It takes ten minutes to walk through with your facilities lead. It saves the kind of week that ends in a compliance call.