Retired 2,000 Handsets Across 25 Hotels. Kept the PBX.
Camila was about to sign off on replacing 2,000 bedside handsets guests had already stopped using.
The landline cost is gone, the PBX stayed, and the guest experience moved onto the phone already in every guest's hand.
Camila is director of guest services for a mid-size hotel chain, about 25 properties, a mix of legacy PBX and cloud infrastructure depending on when each hotel was last renovated.
She did not set out to touch the phone system. She set out to fix a line item that made no sense anymore.
Guests were not calling from the bedside handset. They were calling the front desk from their own phones, ordering room service off a website on the TV, and asking Google for restaurant picks instead of the concierge.
We talked to her about what it took to retire a fixture nobody had questioned in a decade.
What did you do first?
Nothing, for a long time. That was the first mistake. The handsets were not broken, so nobody flagged them. They just sat there, working fine, doing almost nothing.
Every property had one in every room, we paid the phone line for every room, and I would guess less than a tenth of guests ever picked one up except to call the front desk about a lost remote.
I kept telling myself it was not urgent. It is easy to ignore a cost that is not visibly on fire.
What was the breaking point?
The hardware refresh. Our facilities team came to me with a number: over 2,000 handsets across the portfolio were aging out, and the next capital cycle was going to replace them, unit for unit, room for room, with the same thing.
Same wired handset, same phone line, same cost, doing the same almost-nothing. I was being asked to sign off on spending real money to keep something guests had already stopped using. That is the moment it stopped being a "someday" problem.
How did that feel? And commercially?
Honestly, a little embarrassed. Guest services is supposed to know how guests actually behave, and I had let this run on autopilot.
Commercially it was worse than embarrassing. It was a straightforwardly bad trade:
- We were about to spend a hardware budget maintaining a surface that was not earning its keep.
- The thing guests were actually using, their own phone, got nothing from us. No room service ordering, no concierge access, no check-in convenience.
- We were paying twice: once for a landline nobody used, and once in missed guest experience we were not delivering anywhere.
Handsets nobody used
over 2,000 bedside handsets across ~25 properties, a phone line paid per room, used by under a tenth of guests
Hardware refresh forced the question
the next capital cycle would replace 2,000 handsets unit for unit, same cost, same almost-nothing
Kept the PBX, moved to a branded app
Class 5 compatible softphone plugs into the existing PBX like an extension; nothing ripped out
Landline cost gone, experience moved onto the guest's phone
active-user billing at $0.12 per active user on the White Label tier
What changed?
I stopped thinking about it as "replace the phone system" and started thinking about it as "meet the guest on the device already in their hand." That reframe mattered more than any vendor conversation.
Once we found Acrobits, the actual technical piece was almost boring, in a good way. Their softphone is Class 5 compatible, which just means it plugs into our existing PBX the way a phone extension does. The switch does not know or care that the "phone" is now an app instead of a handset.
We did not have to rip out infrastructure that was working fine. The call routing, the engineering underneath, that is the plumbing, and it kept running without us touching it. We only had to change what sat on top.
How they used Acrobits
The guest-facing app went white-label with our logo, our name, on the App Store and Play Store, not "some vendor's app." Inside it, calling still works exactly like the old handset did.
Our own web tools went into Custom Web Tabs. Think of them as browser windows built into the app: room service ordering, concierge requests, mobile check-in, and our loyalty program. None of it is new technology we invented, it is our existing web tools surfaced inside the same app guests use to call the desk. A guest opens one app for everything instead of hunting for a TV menu, a website, and a phone.
Housekeeping alerts came in through the IPC SDK, the piece that lets a web tab talk to the phone itself, not just display a page. Housekeeping used to get room-status updates through a separate handheld system. Now checkout and room-ready alerts arrive as real push notifications inside the same staff app they use for internal calls. The phone buzzes and badges like an actual call came in, because as far as the operating system is concerned, it did. One app, one notification stream, one place to look.
If you had to teach this to someone, what is the one idea?
Unified communications happens in the UI. I used to think "unified comms" meant a big system in a server closet somewhere. It does not.
It happens on the screen the guest or the staff member is actually looking at: calling, room service, concierge, loyalty, housekeeping status, all in the same hand.
The mechanism that makes that possible is web tabs for the guest-facing tools and the IPC SDK for anything that needs to act like a native phone feature. Once I understood that split, I stopped seeing this as a phone project and started seeing it as a guest-experience project that happened to use the phone system as its foundation.
What they needed
- A softphone that plugs into the existing PBX the way an extension does, so working infrastructure stays in place
- A branded guest app on the App Store and Play Store carrying the hotel's name, not a vendor's
- Custom Web Tabs to surface room service, concierge, mobile check-in, and loyalty inside the same app guests call from
- A real bridge to the phone via the IPC SDK so housekeeping alerts buzz and badge like a native call
- Active-user billing so seasonal headcount and occupancy swings never mean paying a flat per-seat tax through slow months
Technical detail
Features that did the work
White-label branding
the guest app and both store listings carry the hotel chain's brand, not Acrobits'.
Custom Web Tabs
existing web tools (room service, concierge, mobile check-in, loyalty) render as native tabs inside the app. One app for everything, not a TV menu plus a website plus a phone.
IPC SDK
web content in those tabs reads call state and fires native notifications. Checkout and room-ready alerts buzz and badge like a real call. Standard web code, native behavior.
Class 5 compatibility
the softphone plugs into the existing PBX like a phone extension, so the switch keeps routing calls and nothing gets ripped out.
Active-user billing
pay per active user, not per seat, so seasonal occupancy and headcount swings do not carry a flat per-seat cost.
And the business?
The landline cost came off the operating bill, one line item, gone, across every property. The PBX stayed, so we did not pay to replace working infrastructure, and the hardware refresh budget that would have gone to another 2,000 handsets went to something that actually moves guest satisfaction.
We also bill on active users, not seats, which matters a lot in hospitality. Headcount and occupancy swing seasonally, and we were never paying a flat per-seat tax through the slow months. On the White Label tier that is $0.12 per active user, so the bill tracks how many people are actually working, not how many badges exist.
There is a broader number worth knowing even though it is not specific to our deployment: branded apps like this typically carry an $8 to $13 per seat per month premium over a plain, unbranded softphone, at 64 to 77 percent gross margin, because the ongoing cost is mostly the same infrastructure. The upside comes from the branding and the experience layered on top, not from rebuilding anything. We have not fully priced that upside into our loyalty and room-service flows yet, but it is the direction this is heading.
A fixture that quietly stopped being used isn't stability.
It's a cost pretending to be infrastructure. Don't wait for the hardware refresh to force the question. Meet people on the device already in their hand, and put the saved budget where it actually moves the guest experience.