Every coverage map is a prediction, and predictions are least reliable at exactly the scale you care about: your house. Here is how to check properly in about ten minutes, using tools you already have, before you commit to anything.
Last updated 31 August 2026
Whenever fixed broadband is missing, late or broken, the answer is almost always mobile. A 5G router, a home hub, a hotspot, a tethered phone: all of them lean on the same thing, which is whether there is a decent mobile signal where the device will actually sit. Get that right and mobile internet is genuinely good. Get it wrong and you have spent money on a box that shows one bar in the kitchen.
The frustrating part is that the obvious way to check, typing your postcode into a coverage map, is the least reliable step in the process. Here is how to check properly.
Coverage maps are not measurements. They are propagation models: software that takes the position and power of each mast, applies terrain and clutter data, and estimates what should arrive at every point on the map. Ofcom's own drive testing has found that operator predictions are reasonable on average but subject to significant error at a local level, which is a polite way of saying the model does not know about the hill behind you, the oak tree at the front, or the foil-backed insulation in your loft.
Ofcom's own checker, Map Your Mobile, blends those predictions with crowdsourced data from real handsets, including how often a connection actually managed 5Mbps. That helps, because real measurements catch what models miss. Ofcom is still explicit that no signal threshold guarantees coverage 100% of the time.
So use maps for what they are good at, which is narrowing four networks down to one or two worth testing, and never as proof.
This is the step most people skip, and it is the one that changes outcomes. Coverage varies enormously between operators on the same street, because they built their masts in different places at different times. Your current network being poor at home tells you very little about the other three.
| Where to look | What it gives you | Worth using because |
|---|---|---|
| Ofcom Map Your Mobile | All four networks side by side, indoor and outdoor, good or variable | Independent, includes real crowdsourced results, one search covers everyone |
| EE, O2, Three and Vodafone checkers | Each operator's own map, usually with separate 5G layers | More current than open data, and shows recent or planned upgrades |
| Neighbours and local groups | Lived experience on your actual street | Beats any model, especially on new estates and in valleys |
If three networks predict variable and one predicts good, you have learned something useful. If all four predict variable indoors, that is a real warning, and no amount of hardware fixes it.
If you take one thing from this guide, take this. Checkers give separate indoor and outdoor predictions, and they routinely disagree. Outdoor is the signal in open air at your postcode. Indoor is what is left after that signal has fought its way through a wall.
An address can genuinely show good outdoor 5G and variable indoor 5G at once. Both are true. The map is not contradicting itself, it is telling you the building is the problem. If the router will sit on a desk inside the house, the indoor row is the only one that matters.
The categories are worth translating too. "Variable" means voice, texts and lower quality video are likely to work. "Good" means streaming and data heavy use should hold up. For a household connection, variable is not a near miss. It is a warning.
Ten minutes at the property beats every map. If you can get inside the house, or you have a friend or an agent who can, do this.
Signal strength readings are negative numbers in dBm, and closer to zero is stronger. As a rule of thumb:
| Reading (RSRP) | What it means | Realistic expectation |
|---|---|---|
| -80 dBm or better | Strong | Fast and stable, close to the best that mast can give you |
| -80 to -95 dBm | Good | Comfortable for streaming, video calls and a full household |
| -95 to -105 dBm | Workable | Usable, but expect it to wobble at busy times, and placement matters |
| -105 dBm or lower | Weak | Marginal. Fine for messages, unreliable for work |
Not all 5G is the same, and this explains most of the confusion people run into. Most UK 5G capacity sits on mid band spectrum around 3.4 to 3.8GHz. It is fast and it carries a lot of traffic, but higher frequencies fade quickly through walls, modern coated glass and metal. Low band 5G at 700MHz behaves in the opposite way: slower, but it travels much further and gets inside buildings far more reliably. Networks use it as their coverage layer precisely where the mid band cannot reach.
The practical consequence is that "5G is available here" can mean two very different experiences. Fast 5G in the garden and 4G in the back bedroom is not a fault, it is physics. It is also why hardware choice matters: a router that supports the low band your network uses locally will hold a connection in places where one that does not simply cannot.
Where this leaves you. Checking coverage properly is worth the effort, but it still ends in a probability rather than a guarantee, and you usually need to be online before you can be completely certain.
That is the risk we deliberately took onto our side of the table. Airgen Go rents pre-configured unlimited 5G routers from £39 a week, and before every dispatch we check your postcode across the UK networks and fit the SIM predicted strongest at your address, rather than handing you whichever network we happen to buy from. If the signal disappoints anyway, send it back within 48 hours for a full refund. No credit check, no direct debit, no contract to unwind.
Marginal is common, and it is often fixable. Work through these in order, because they run cheapest first.
In solid 5G coverage, expect roughly 100 to 300Mbps, with the average for UK 5G home broadband sitting near 150Mbps. On 4G fallback, expect roughly 20 to 80Mbps. Both are comfortably enough for 4K streaming, video calls and a house full of devices. For context, most video calls need under 5Mbps and 4K streaming needs around 25Mbps.
The honest caveat is variability. A mast is shared capacity, so speeds move around through the day in a way a good full fibre line does not, and evenings are busiest. Upload is usually lower than download, which matters if you back up large files. For bridging a gap that trade is easy. As a permanent fibre replacement, judge it on the evening numbers, not the ones you measured at 11am on a Tuesday.
Check every network rather than just yours, read the indoor prediction rather than the outdoor one, and spend ten minutes taking real readings where the router will actually sit. That costs nothing and beats any single map. Then favour whichever option you can reverse cheaply, because no check is certain until the box is switched on in the room.
They are predictions from propagation models, not measurements at your door. Ofcom's drive testing finds operator predictions reasonable on average but subject to significant local error, because terrain, trees, neighbouring buildings and wall construction all change the result. Use maps to shortlist networks worth testing, then confirm with a real device at the address.
Outdoor is the predicted signal in open air at your postcode. Indoor is what remains after that signal passes through a typical building, which weakens it considerably. An address can show good outdoor 5G and variable indoor 5G at the same time. If the router lives inside, the indoor layer is the one that matters.
On iPhone, dial the field test code (star, 3001, hash, 12345, hash, star) and press call, then read the RSRP figure. On Android, look under Settings, About phone, then SIM or Network status, or use a free app such as Network Cell Info Lite, which also shows the band. Readings are negative dBm values, so closer to zero is stronger. Around -80 dBm is strong, -100 dBm is workable, -110 dBm is weak.
Most UK 5G capacity sits at 3.4 to 3.8GHz, which is fast but fades quickly through walls, coated glass and metal. Low band 5G at 700MHz travels much further and penetrates buildings better, but it is slower and not deployed everywhere. That gap is why signal can be strong in the garden and drop to 4G in a back bedroom.
Roughly 100 to 300Mbps in solid 5G coverage, with the UK average for 5G home broadband around 150Mbps, and roughly 20 to 80Mbps on 4G fallback. Expect more variation through the day than a fixed line, because mast capacity is shared. Anything above about 30Mbps is comfortable for a normal household.
Move the router upstairs and near a window first, since height and glass often fix it. Check whether another network is predicted stronger at the same postcode. If it is still poor, avoid anything with a contract or credit check attached and choose an option you can return quickly instead.
Also worth reading: how long broadband installation really takes by provider, how to get WiFi with no landline, and what to do when your broadband goes down.
Unlimited 5G WiFi from £39, delivered next working day, with the SIM matched to the strongest network at your postcode. No credit check, no contract, and a 48 hour money back guarantee if the signal disappoints.
See pricing →