Coverage, honestly

PoC coverage: how 4G/5G push-to-talk really performs

The honest version: with the right SIM, PoC coverage beats any single mobile phone and reaches places no conventional radio system could economically touch — and it still fails in exactly the places your phone does. Here's how to know which you'll get, before you sign anything.

The SIM decides more than the handset

A PoC radio's coverage is exactly the coverage of whatever networks its SIM can use — so the SIM choice is the coverage decision. A single-network SIM gives you one operator's map, holes included. A multi-network roaming SIM can attach to any of EE, O2, Vodafone or Three, and its practical coverage is the union of all four maps — which is why a well-specified PoC fleet routinely out-covers every phone in the room.

The trap inside that promise is steering. Many roaming SIMs are steered: they prefer a home network and only look elsewhere when it disappears entirely, so a radio can sit on one bar of the preferred network while a rival's full-strength mast overlooks the same yard. An unsteered (or "non-steered") SIM attaches to the strongest signal, full stop. When a fleet "has coverage problems" that the coverage maps say shouldn't exist, a steered SIM is the first thing a practitioner checks — and the question to put to any provider in writing is: "Is the SIM unsteered, and which networks does it roam across in the UK?" The full handset→SIM→platform chain is explained in how PoC networks work.

Indoors: the reality and the Wi-Fi rescue

Mobile signal weakens indoors — modern insulated buildings with foil-backed board and coated glass can be surprisingly hostile, and basements, plant rooms and lift cores are often dead outright. Three things decide whether this matters for PoC:

Rural notspots — and why checkers mislead

The UK's remaining coverage gaps are rural: valley floors, moorland, forestry, stretches of minor road. A multi-network SIM shrinks these dramatically — the operators' rural builds don't overlap neatly, so the union of four maps is much stronger than any one — but it cannot cover a location no operator reaches.

Assess this honestly, and be sceptical of the online coverage checkers. They are predictions from terrain modelling, not measurements, generally assume outdoor use at roughly head height or above, and are least accurate precisely at the margins — the valley farm, the site cabin behind a hill — where your decision actually lives. Use checkers to shortlist, then prove the marginal sites with real handsets. In genuinely uncovered terrain, the choice is conventional radio for local work or accepting store-and-forward behaviour: PoC radios reconnect automatically and queue their GPS reports, but live voice needs live data.

The latency truth: about half a second

Press the button on a conventional radio and you're transmitting immediately. Press it on a PoC radio and your audio is compressed, sent over the network to the platform, and relayed to the group — typically 300–600 milliseconds end to end. That half-second is real and permanent; no handset or tariff removes it. In practice teams adapt within a day — the habit is pressing, pausing a beat, then speaking, and most platforms play a "granted" tone that trains it automatically. Where it genuinely matters — crane lifts, banksman work, race marshalling — teams either accept the beat, keep conventional radio for the tight loop, or run both. What latency does not do is degrade audio: on a healthy connection PoC voice is consistently clearer than analogue radio at range, with none of the hiss and fade.

What a proper pilot measures

Every credible provider will lend five to ten radios for a month. The difference between a pilot and a demo is that a pilot is designed to fail — you send the radios to the worst places you operate and write down what happens. Measure:

TestHowPass looks like
Worst-location voiceDaily calls from the basement, the far yard, the rural depot, the known dead spotsCalls connect first press, audio intact — or a documented, acceptable exception list
In-vehicle at speedMotorway and back-road runs on your real routesNo dropped talk-bursts during network handovers
Wi-Fi handoverWalk from yard to basement mid-conversationSwitch to Wi-Fi without a dead gap that matters operationally
Failure behaviourDeliberately visit a notspot, then leave itRadio reconnects itself; queued GPS arrives; no user fiddling
Latency toleranceGive it to your most impatient supervisor for a weekNobody mentions the delay by day three

Keep a simple log — location, time, result — because "it seemed fine" is not data, and the log becomes your acceptance criteria for the contract. A provider reluctant to support a worst-case pilot has answered your coverage question already.

Versus DMR range: the physics in one paragraph

Conventional radio range is physics you own: UHF handhelds manage roughly 0.5–2 km to each other through buildings and terrain, and a well-sited repeater lifts a site or a town into coverage — beyond that you're buying more repeaters. PoC range is geography you rent: effectively national wherever any partner network or Wi-Fi exists, with zero infrastructure, and precisely zero where none does. Neither is "better range" in the abstract — one is short, private and self-contained; the other is vast, shared and network-dependent. Which failure mode you can live with is the real question, and the cost side of the same decision is worked through in the total cost of ownership guide.

Rule of thumb: if your phones broadly work where your people work, an unsteered multi-network SIM will out-cover them — it sees four networks, your phone sees one. Where phones are dead across the operating area, no brochure changes the physics: pilot first, hybrid if the pilot says so.

FAQ

How good is PoC coverage in the UK?

On an unsteered multi-network SIM: better than any single phone, because the radio uses the strongest of EE, O2, Vodafone or Three at each location. Genuine notspots — deep rural terrain, uncabled basements — remain, and only testing proves them either way.

What's the difference between steered and unsteered SIMs?

Steered SIMs cling to a preferred network even when it's weakest; unsteered SIMs attach to the strongest available. For PoC, insist on unsteered — it's the biggest single coverage variable.

How much delay does PoC have?

About half a second press-to-voice (300–600 ms). Teams stop noticing within a day; time-critical marshalling work should test it deliberately.

Can I rely on operator coverage maps?

Only for shortlisting. They're modelled predictions, weakest at the exact margins that decide radio projects. Prove marginal sites with handsets during a pilot.

Next: Migrating from analogue or DMR to PoC — the honest guide →