As of October 2024, about a third of American workers whose jobs can be done from home say they work from home all the time, and 43% now keep a hybrid schedule, according to Pew Research Center’s five-year look back at how the pandemic reshaped US workplaces. The exact number matters less than what it does to a phone system. A desk phone is a fixed endpoint bolted to a fixed desk. When the desk is occupied two or three days a week, or not at all, the handset stops being the place work happens and starts being a line item.
That is the plain reason businesses are switching to softphones, and it is worth saying out loud before the marketing versions take over. Most switching content promises to halve your phone bill. The honest version is messier: the license usually costs about the same, the hardware saving is real but small, and the money that actually moves is the labor you spend touching phones.
This guide walks through what changed, the five reasons teams give when they move off desk phones, a three-year cost model with every assumption on the table, what a softphone migration involves when you are replacing an app you already own, and the part nobody publishes, which is what switching does not fix.
What a Softphone Is, and What Sits Under It
Four definitions first, because half the arguments about softphone vs desk phone are really arguments about vocabulary.
Softphone. A software application that places and receives phone calls over an internet connection. It runs on a laptop, a mobile phone, a tablet, or a browser tab. It registers to your phone system the same way a physical handset does, using the same credentials and the same protocol.
Desk phone (or hard phone). The physical unit on the desk, usually from a vendor like Poly or Cisco, cabled to the network and provisioned to one extension. It does one job, it does it without a computer, and it keeps doing it when the person is not there.
Hosted PBX (or cloud PBX). The phone system itself, running in a provider’s data center instead of a closet in your building. It holds your extensions, your call routing, your voicemail boxes, your hunt groups, and the numbers assigned to them. UCaaS, unified communications as a service, is the wider bundle that puts calling, video, messaging and meetings on one subscription.
Total cost of ownership (TCO). Everything the phone system costs over a defined period, not just the monthly bill: hardware, provisioning labor, moves and changes, replacements, support time, and the cost of the switch itself.
Here is the part that gets lost. Moving to a softphone does not change your PBX, your numbers, or how you are billed. In the channel, PBX providers and managed service providers license digital numbers to businesses and bill by seat or by concurrent session, and for two decades the delivery mechanism at the end of that chain was a handset on a desk. The seat is still a seat. What changes is the thing the seat rings on.
Why the Desk Phone Stopped Making Sense
The desk phone did not get worse. Its assumptions did.
It assumes one person, one desk, one location, all day. It assumes the phone is a separate tool from the computer, so nothing that happens on the call needs to know anything about the customer record open on the screen. It assumes that adding a person means shipping a box, and that moving a person means somebody walks over with a screwdriver.
Labor statistics on telework put the share of people at work who telework or work at home for pay at 22.9 percent in the first quarter of 2024, up from 19.6 percent a year earlier, in a Bureau of Labor Statistics review of telework trends. Roughly a fifth to a quarter of the working population is now regularly somewhere other than the desk the handset is sitting on. A phone that cannot follow them is not a phone system, it is furniture with a dial tone.
The second thing that broke is integration. A desk phone does not know who is calling beyond what the carrier passes. It cannot open a ticket, log a call against a deal, or fire a webhook. Everything a modern support or sales workflow wants from a call happens somewhere the handset cannot reach. That is the practical case for replacing desk phones in any team where the call is the start of a process rather than the whole of it, and it is worth reading up on connecting a softphone to your CRM before you assume the integration will just be there.
The Five Reasons Businesses Actually Switch
These are the reasons that survive contact with a budget conversation. Each one stands on its own.
- Cost, but not where the brochures say. The seat license rarely drops much when you switch to a softphone. What drops is capital hardware and the hours spent provisioning, shipping, replacing and re-siting it. In a stable single-site office with low turnover, that saving is modest. In a business with three sites, 30% annual churn and a habit of reshuffling desks, it is the whole case.
- The work moved and the phone did not. Hybrid and remote schedules made the fixed endpoint the wrong endpoint. A softphone puts the business number, the business voicemail and the business caller ID on whatever device the person already carries, which is the single most common trigger for business softphone adoption we hear about.
- The phone system had to talk to everything else. CRM screen pops, ticket creation, click-to-dial from a record, call logging, transcription. None of that is available to a handset. All of it is available to an application that lives on the same machine as the browser.
- Scaling stopped being a shipping problem. Adding 12 people to a desk-phone estate means buying, configuring, shipping and racking 12 devices, and hoping they arrive before the people do. Adding 12 people to a softphone estate is 12 rows in an admin console.
- Feature velocity. Firmware on a hard phone updates rarely and awkwardly. An application updates on its own release cycle, so messaging, video, presence and call controls arrive without anyone touching an endpoint.
None of the five is a knockout argument on its own. Together they explain why so many teams end up moving to a softphone even when the monthly line on the invoice barely moves.
The Cost Case, Worked: Three Years, 25 Seats
Most cost claims about switching to softphones are unsourced. So here is a model instead of a claim, with every input visible so you can replace it with your own quotes. These are illustrative planning assumptions, not measured figures from any particular deployment, and the point of the exercise is the shape of the answer rather than the total.
Assumptions used in both columns: 25 seats, 36 months, a single office, an internal or contracted technician rate of $75 per hour, and a hosted PBX seat license that already exists. Handset price assumed at $150 for a mid-range SIP desk phone. Headset price assumed at $70. Seat license assumed at $20 per seat per month on the desk-phone path and $22 on the softphone path, because a softphone plan usually carries messaging and video that a bare SIP seat does not.
| Cost line (3 years, 25 seats) | Desk phone path | Softphone path |
|---|---|---|
| Endpoint hardware | $3,750 (25 handsets at $150) | $1,750 (25 headsets at $70) |
| Initial provisioning labor | $940 (30 min per device) | $310 (10 min per user) |
| Seat licenses | $18,000 ($20 per seat per month) | $19,800 ($22 per seat per month) |
| Moves, adds and changes | $900 (5 changes per year at $60) | $450 (2 admin hours per year) |
| Replacement and failure | $900 (approx. 8% of devices per year) | $770 (approx. 15% of headsets per year) |
| Cabling and power at new desks | $400 (8 desks at $50) | $0 |
| Three-year total | $24,890 | $23,080 |
The gap is $1,810 over three years, or about 7%. That is a real saving and it is nothing like 50%.
Now change one assumption. Run the same model across three sites with people moving desks quarterly, and the moves-and-changes line on the desk-phone side stops being $900 and starts being the largest variable cost in the table, because every change is a person walking to a desk. Run it in a business that hires and loses 40% of its headcount a year, and the hardware and provisioning lines move with it. Run it in a stable 12-person office where nobody moves and nobody leaves, and the desk-phone path can genuinely come out ahead once you account for the headsets you now have to buy and replace.
That is the useful conclusion about softphone vs desk phone economics: the license is a wash, and the saving lives entirely in how often your organization touches a phone. Count that number first. If you are also weighing whether to build your own client rather than buy one, the arithmetic is a different exercise again, and what building one actually costs is where that comparison belongs.
Softphone vs Desk Phone vs Mobile-Only: An Honest Comparison
A third option exists and gets ignored: skip the phone system, give everyone their mobile number, and be done. It works for some businesses and fails badly for others. Here is the comparison including what each option costs you.
| Consideration | Desk phone | Softphone | Mobile-only |
|---|---|---|---|
| Works during a power cut | Yes, with PoE and a UPS | Only if the device and network stay up | Yes |
| Company owns the number | Yes | Yes | Usually no, the employee does |
| Call recording and compliance | Handled by the PBX | Handled by the PBX or the app | Hard to do lawfully and consistently |
| Reception console and busy lamp | Mature, physical keys | Software equivalent, needs training | Not available |
| Audio consistency | Very predictable | Depends on device, headset and network | Depends on cell coverage |
| Analogue devices (fax, alarm, lift line) | Supported via ATA | Not supported, needs a separate path | Not supported |
| Cost of adding a person | Buy and ship a device | Add a license | Nothing, and that is the problem |
| Cost of losing a person | Recover the device | Revoke the license | The number leaves with them |
| CRM and workflow integration | Minimal | Native | Minimal |
The line that ends most mobile-only conversations is number ownership. If the customer relationship lives in a phone number that belongs to a departing salesperson’s personal account, you do not have a phone system, you have a liability.
What Small Teams Should Evaluate Before Switching
The evaluation questions that actually predict whether a softphone migration goes well are unglamorous. Work through these before you compare feature grids.
- Does it register to the PBX you already have? This is the first question, not the last. If the application speaks SIP to your existing hosted PBX, you are changing one layer. If it requires its own back end, you are changing your phone system, your numbers and your billing at the same time.
- What happens to mobile calls when the app is closed? Push notification behavior, background wake and battery handling are where cheap softphones fall apart. Test an inbound call to a phone that has been in a pocket for four hours, not a phone you just unlocked.
- Which codecs, and can you prove the audio? HD voice on paper is not HD voice on your network. Run a pilot on the actual office connection and the actual home connections.
- How are users provisioned? Manual SIP credential entry for 200 people is a project. Bulk provisioning, SSO, or a QR-based enrollment turns it into an afternoon.
- What is the admin console allowed to do? If a desk move, a new hire or a call-flow change still needs a support ticket to the vendor, you have not removed the labor cost that justified the switch.
- What is the porting story? If you are moving to a softphone on the PBX you already have, there may be no port at all. If you are changing carriers, get the porting timeline and the required documentation in writing before you sign.
- If you resell, whose brand is on it? For a service provider, an unbranded or vendor-branded app on a customer’s phone is a monthly reminder that the customer’s relationship is with someone else.
A fuller version of this list, organized as a buying process rather than a checklist and written for teams part-way through switching to softphones, is set out in our guide on choosing a softphone for a business, and the underlying distinction between the app and the transport is covered in how softphones and VoIP fit together.
Replacing a Softphone You Already Have
A large share of switching traffic is not about replacing desk phones at all. It is people who already run a softphone and want a different one, usually after a price change, an interface redesign, or a platform version they cannot get support for.
This is where we learned something that changed how we talk about the product. When we asked one customer how they found us, the answer was that they had gone to ChatGPT and asked for softphone alternatives to Acrobits that connect with NetSapiens. The interesting part is the second half of that question. They were not asking a softphone vs desk phone question at all, and they did not ask for the best softphone. They asked for a softphone that would work with the PBX they intended to keep. Replacement demand is constrained by the back end, not by the feature grid, and almost nobody writes for that constraint.
That constraint is also why our own softphone is built the way it is. The iotum Communicator is layered on top of the NetSapiens PBX rather than replacing it, so a service provider swaps the end-user experience while the platform and the customer’s numbers stay exactly where they are, with no rip and replace. The switch that gets approved is almost always the one that changes the fewest layers.
If you are replacing an existing softphone rather than moving to a softphone for the first time, the migration itself breaks down like this.
- Inventory the registration details first. SIP username, password, domain, transport, port, and codec list for every extension. If you cannot export them, budget for retyping them.
- Pilot on two extensions for a week. Not a demo call. A real week, including a call taken in a parking lot and one taken on a hotel Wi-Fi network.
- Run both applications in parallel. Most PBX platforms allow multiple registrations per extension. Overlap removes the cutover cliff.
- Map the DIDs, not just the users. Shared lines, hunt groups, the main number and the after-hours path are where cutovers go wrong.
- Accept what does not come with you. Call history, in-app message threads, and locally stored voicemail generally stay behind. Contacts usually migrate. Recordings live with the PBX or the recording platform, so confirm retention before you touch anything.
- Decide the porting question early. Number portability is the process of moving your existing phone numbers between carriers so the number stays yours when the provider changes. If you are only changing the application on top of an unchanged PBX, no port is required and the whole timeline collapses.
Done this way, replacing a softphone is two weeks of careful work rather than a quarter-long project.
Why VoIP Providers and MSPs Launch Their Own Branded Softphones
There is a second audience for all of this, and it is the one iotum actually sells to: the VoIP service providers, ITSPs, MSPs and carriers who resell communications rather than consume it.
For them, business softphone adoption among their customers is a revenue conversation, not an IT one. A branded softphone raises average revenue per user, because messaging, video and meetings become part of the seat instead of a competitor’s separate subscription. It reduces churn, because the customer’s daily communications tool carries the reseller’s logo rather than a vendor’s. And it protects the relationship, because when the customer looks at the app on their phone they are reminded who they buy from.
The alternative is building one. That is a genuine option and a genuine budget. Our own view, formed from years of these conversations, is blunt: building a softphone costs real money and a white-label license usually does not, so unless the client itself is the product you are selling, buying the front end and keeping your platform is the cheaper path. We package the same underlying stack as a white-label suite for service providers, which is the version of this most resellers actually deploy.
The wedge, in practice, is rarely the platform. When a reseller’s customers start switching to softphones, the thing being replaced is the end-user application sitting on top of the platform, not the platform itself. Plenty of resellers are perfectly happy with their back end and unhappy with the client their users open every morning, which is precisely the gap a front-end swap fills.
What Switching to Softphones Does Not Fix
This is the section most vendors leave out, so here it is plainly.
It does not fix bad audio caused by your network. Jitter, packet loss and an oversubscribed office connection sound exactly the same through an application as they did through a handset. If calls are bad today, fix QoS and the uplink before you blame the endpoint.
It does not fix call routing you never designed. Switching to softphones leaves the call plan exactly as it was. A confusing IVR, a hunt group that rings the wrong four people, and a main number with no after-hours path stay confusing after the switch. Those live in the PBX, and the PBX is the layer you are keeping.
It does not fix people who do not want to be reachable. Presence and status make availability visible. They do not create it.
It does not solve analogue endpoints. Fax machines, elevator emergency lines, alarm panels and door entry systems need an analogue adapter or a separate service. Plan for them explicitly, because they are the classic Friday-afternoon discovery.
It does not survive a power cut on its own. A desk phone on PoE with a battery-backed switch rides out a short outage. A laptop softphone rides out exactly as long as the laptop, the Wi-Fi access point and the internet connection do. If continuity matters, keep a mobile fallback path and test it.
It does not remove the training. A softphone migration hands people a new interface, and a reception console replaced by software needs an hour with the receptionist, not a link to a help article. Transfer, park, pickup and busy lamp behavior all move, and the people who use them fifty times a day will notice on day one.
Being honest about these is not an argument against switching to softphones, and it is not a reason to stay on desk phones. It is the difference between a switch that lands and one that gets reversed in month three.
Frequently Asked Questions
Is a softphone cheaper than a desk phone?
Usually yes, but by less than most softphone vs desk phone comparisons claim. In the three-year model above, a 25-seat office saved roughly 7%, because the seat license dominates the total and does not fall much. The saving grows with the number of sites, the rate of staff turnover, and how often desks get reshuffled, since those are the activities that cost hardware and technician hours.
Can you keep your existing phone numbers when switching to a softphone?
Yes. If you are moving to a softphone on the same PBX and carrier and only changing the application, nothing moves at all and no port is needed. If you are also changing carrier, the numbers port across through the normal number-portability process, which typically takes a few business days for a handful of numbers and longer for larger or geographically mixed ranges.
Do softphones work without a dedicated VoIP provider?
A softphone needs something to register to. That can be a hosted PBX, an on-premise PBX, or a SIP trunk from a carrier, but it cannot make external calls on its own. The only exception is an app that includes its own calling service, in which case the provider is bundled rather than absent.
What do teams lose when they drop desk phones?
Teams replacing desk phones give up four things, reliably: power-cut resilience, analogue device support for fax and alarm lines, the physical reception console with its busy lamp keys, and the predictability of one device doing one job. Each has a workaround. None of them is free, and all four should be on the plan before the handsets are recycled.
How long does a softphone migration take?
For a single-site business under 50 seats that is keeping its PBX, plan two to three weeks: one week of piloting, one week of parallel running, and a short cutover. Changing carrier or PBX at the same time roughly triples that, mostly because of porting windows and call-flow rebuilds.
Key Takeaways
- The license is a wash, the labor is the saving. Switching to softphones moves money out of hardware and technician hours, not usually out of the monthly seat price.
- Count how often you touch a phone. Sites, turnover and desk moves predict the size of your saving better than headcount does.
- Replacement demand is constrained by the back end. Buyers replacing an existing softphone are looking for something that works with the PBX they intend to keep.
- The switch that gets approved changes the fewest layers. Swapping the front-end application while the platform and numbers stay put is faster, cheaper and less risky than a full platform migration.
- Plan for the four losses. Power-cut resilience, analogue endpoints, the reception console and audio predictability all need an explicit answer before cutover.
- For resellers, the branded app is the relationship. A white-label softphone raises revenue per seat and keeps the customer looking at your logo rather than a vendor’s.
The businesses that get this right treat it as an endpoint decision, not a platform decision. They keep the PBX, keep the numbers, change the thing people actually open, and spend the saved effort on the call flows and the training instead.
If you are a VoIP service provider, ITSP or MSP whose customers keep asking about switching to softphones, the iotum Communicator is the white-label softphone we build for you to resell under your own brand, sitting on top of the PBX you already run. Book a demo and we will walk through it with your stack in front of us.