The easiest way to choose between PoC and conventional radio is to compare two handsets, notice that both have a PTT key and a speaker, and pick the cheaper one. That decision usually looks fine in month one and uncomfortable by year three, when the recurring line items have accumulated or when coverage has to be explained to operations.
The two technologies fail differently, cost differently and are administered by different people. This article sets out those differences in the terms a procurement or operations team can act on, and ends with a decision table that can be worked through with a supplier.
What does PoC actually change for the user?
PoC — push-to-talk over cellular — keeps the interface users already know: a handheld with a PTT key, talkgroups and a loud speaker. What changes is the plumbing. Voice is digitised and carried as data over a mobile network to a platform, and the platform distributes each transmission to the rest of the group. A radio channel is replaced by a subscription and a server.
That single change has three consequences a buyer should feel immediately. Coverage stops being a radio engineering problem and becomes a mobile-coverage question: with a data plan, a PoC fleet works wherever the carrier network works — across a city, a region or a country — without building a site. Group management becomes software: dispatchers can create and reassign groups, watch positions on a map, replay recordings and send messages without touching the radios. And administration becomes an IT task — SIMs, accounts and platform users — rather than a programming bench task.
Conventional DMR runs on the opposite principle. DMR Tier II is the conventional, non-trunked mode of the DMR standard: 12.5 kHz channels and two-slot TDMA, so one repeater carries two independent talk paths on a single frequency pair. Voice travels over licensed radio spectrum through infrastructure the operator owns, and coverage is what the site, the antennas and the terrain deliver. Crews can also talk directly, radio to radio, without the repeater at all at short range — a resilience property that is easy to forget until the day it matters.
How do the two compare on coverage?
PoC coverage is national on paper and very uneven in practice. The limits are the familiar ones: basements, plant rooms, thick concrete, steel racking in large warehouses, tunnels, remote fields, and anywhere the network is congested or the signal is only just usable. A PoC fleet cannot close those gaps the way a radio system can. It either waits for the carrier to improve coverage, adds indoor coverage of its own, or accepts the dead spots. In exchange, PoC needs no radio sites at all, which is decisive for fleets spread over a wide area and for temporary operations — an event, a project, a seasonal site — that cannot justify infrastructure.
Conventional coverage is bounded, but it is knowable and repairable. A repeater on a suitable roof or mast covers what its height, the terrain and the buildings allow; if a zone is not covered, a second site or a different antenna arrangement closes the gap. Nothing outside the operator control sits between the user and the conversation, which is why conventional radio is still specified for plants where communications must keep working through a regional power or network event. Direct-mode simplex adds a final layer: if the repeater or its power fails, crews inside a building or a yard can still reach each other, over a much shorter range.
The practical summary: PoC removes the site-building problem and adds a dependency; conventional radio removes the dependency and adds a site-building problem. Where a fleet operates across many kilometres and hires and loses staff constantly, PoC usually wins on coverage and speed. Where an operating team lives inside a fixed perimeter all day, a conventional system with a well-placed repeater gives more coverage per unit of capital and per minute of downtime.
Where does the money go over five years?
The two options do not just have different prices; they have different cost shapes, and the shape matters more than the unit price in any five-year picture.
PoC front-loads very little. The capital cost is the terminals plus chargers, vehicle kits and accessories. Almost everything else is recurring: a cellular data plan for each SIM, sized by how much the fleet actually talks and whether the platform adds photos or video; a platform or subscription fee per terminal or per user; and the cost of hosting or integrating whatever the platform does not include — a dispatch console, recording storage, integration with existing systems. Add the administrative effort of SIM and account management, which is real labour even when it never appears on an invoice. The defining property of this structure is that recurring cost scales with headcount and never stops: a fleet of fifty users costs less per month than a fleet of five hundred, but both are still paying in year five exactly what they paid in year one.
Conventional systems front-load more. The terminals are one line item; the site work is usually the bigger one — repeater, antenna and feeder, combiner and filtering if several channels share an antenna, power and backup, lightning protection and installation labour, plus a frequency licence applied for in the destination market, with fees and conditions set by the national regulator. Maintenance — site visits, spares, battery replacement, occasional reprogramming — runs for the life of the system. Against that, voice traffic carries no per-minute or per-user charge, and the infrastructure cost is shared by every user, so cost per user falls as the fleet grows and stays low for as long as the fleet lasts.
Which is cheaper depends on the fleet. Small, spread out, seasonal or constantly changing fleets usually find PoC cheaper and faster to stand up. Large, stable fleets concentrated on one or two sites usually find a conventional system cheaper over five years, because the recurring bill never starts. Cellular plans, platform terms and site builds vary by market, quantity and configuration, so both options should be quoted against your own fleet size, talk profile and site plan rather than compared on list price.
How much latency and reliability should you expect?
On a conventional radio, pressing the PTT key opens the transmitter immediately, and audio starts as soon as the channel is free. Users hear each other with so little delay that normal radio discipline — short transmissions, immediate acknowledgement — works naturally. That immediacy is not a feature; it is a property of putting voice straight onto a radio channel.
PoC has more work to do before anyone hears anything. The key press becomes a request that travels over the mobile network to the platform, which then distributes the audio to the rest of the group. That round trip adds a noticeable wait before the floor opens — on the order of a second or more under good conditions, and longer when the network is busy, when the terminal sits at the edge of coverage or when the platform is distant. Mobile data networks are built for general traffic, with user-plane latency measured in tens of milliseconds on a healthy cell, but they do not by default give priority to PTT traffic, and the platform and audio path add their own time. The result is a system that is perfectly usable for routine coordination and feels sluggish for the fast, interrupt-driven exchange a radio excels at.
Reliability is a different question. A conventional repeater keeps working as long as it has power and its links are intact, and its capacity is reserved for its own users. A cellular network is shared with everyone in the area: a match, a festival, a storm or a local outage can degrade or stop PoC communications for a whole fleet at once, and handovers between cells can interrupt audio briefly even in good conditions. Indoor penetration is the most common surprise in the first month.
What does running a fleet of SIMs involve?
A PoC fleet is a subscription fleet, and the subscription has an operational life of its own. Each terminal needs a SIM with a data plan sized to its usage, usually on a machine-to-machine tariff with its own activation and termination rules; many deployments use a private access point name so that traffic stays on a controlled path. Some markets need multi-carrier arrangements or roaming agreements because a single carrier does not cover everything the fleet does.
Day to day, someone has to own the SIM list: activating new cards, deactivating cards when staff leave or a contractor finishes, watching data consumption so that one heavy user does not exhaust a pooled allowance, replacing broken or lost SIMs, and keeping the platform user list aligned with the payroll list. On the platform side, accounts, group definitions, geofences and firmware updates become recurring housekeeping, and every change is a chance for the fleet and the platform to disagree about who is where. None of this is difficult, but it is work, and it stops the moment someone stops doing it.
The conventional equivalent is smaller in volume but has its own tasks: programming and codeplug management, keeping a record of which radio carries which channels, renewing the licence, maintaining the site, and verifying the system after any change. There is no monthly traffic bill, and adding a user means adding a radio rather than opening a subscription — a difference that matters most when a fleet grows quickly.
When is a converged dual-mode terminal the right answer?
Dual-mode terminals carry both a public-network PTT application and a conventional radio module in one handset, and some add dual SIM and a separate PTT key for each mode. They cost more than a PoC-only radio, and they exist because three real situations keep appearing.
The first is a coverage hole that neither option covers alone: a site where cellular coverage fails inside the buildings but a conventional channel reaches everywhere, or a fleet whose vehicles travel far beyond radio coverage yet must still talk to a fixed site. The second is migration. A converged terminal lets an operator move from conventional to PoC gradually — running the platform alongside the existing channels, proving the level of service during real operations, and retiring the radio layer when the evidence supports it rather than when a contract ends. The third is contingency: keeping a conventional channel available for the moment the public network is congested, degraded or out of service, using the same device, which is far more likely to be understood under pressure than a second radio kept in a cupboard.
Two management points decide whether converged terminals work in practice. Users have to be trained on which mode and which key serves which purpose, or someone will transmit on the wrong system at the wrong moment. And the two identities — a number and a subscription on one side, a programmed channel plan on the other — both have to be administered, so the operational workload is the sum of both worlds rather than a replacement for one of them. In this range, BF-SCP950 and BF-TD910U are the converged options, BF-TP800 gives each mode its own PTT key, and BG530 is a cellular-only terminal for fleets that do not need a radio layer.
So which one should you choose?
Work down the table with the people who will operate the system, not only with the people who will pay for it. A single honest row — for example, a warehouse with poor indoor mobile coverage — can settle the whole question.
| Decision criterion | Leans PoC when | Leans conventional radio when |
|---|---|---|
| Spread of the fleet | Users are spread across several towns, a region or a country | Users work inside one site, campus or industrial park |
| In-building and underground coverage | Carrier indoor penetration is known to be adequate for the work | Coverage has to be engineered, or indoor areas are known weak spots |
| Infrastructure on site | There is no mast, cabinet or power to build a radio site | Antennas and equipment can be hosted, powered and maintained |
| Time to service | The fleet has to be talking within weeks | A site survey, installation and commissioning schedule is acceptable |
| Budget shape | Low capital cost matters, and a monthly cost per user is acceptable | Capital is available, and a recurring per-user fee is unwelcome |
| Fleet stability | Headcount changes often, or the operation is seasonal or project based | The team is large and stable over several years |
| Latency sensitivity | Routine coordination, where a short wait before speaking is tolerable | Work is directed in real time and a delay changes how crews operate |
| Resilience requirement | Dependence on the public network is acceptable for the level of service | Communications must survive public network congestion or an outage |
| Licences and administration | You prefer not to hold a radio frequency licence | Spectrum licences can be obtained and administered locally |
| Data and voice handling | A hosted platform, cloud recording and mobile dispatch are acceptable | Voice and records have to stay inside your own radio system |
Most real deployments end up as a mix rather than a clean choice: PoC for the wide-area and administrative layers, conventional radio for the sites where coverage and immediacy decide whether the work goes well. If the decision is still open, the cheapest way to close it is a two-week pilot with one team and one site: measure the wait before the floor opens, log the dead spots, watch battery drain over a full shift, and count the hours of SIM and platform administration. Request quotations for both architectures against those real numbers, because carriers, platform terms and site builds differ enough between markets that both options have to be quoted per configuration rather than compared as list prices.