5G coverage in Belgium: your municipality's figure, not the country's

"5G covers 95 % of Belgian households." The figure is accurate, it comes from the BIPT, and it is of no use to you. You do not live in 95 % of Belgium: you live on a street, in a municipality, and that is the number to go and find before signing anything.

The good news is that it exists. The regulator publishes an atlas searchable by postcode and by street, with four signal levels and three different denominators. Almost nobody reads it properly, including the comparison sites that copy a national percentage eighteen months out of date.

Why do national percentages tell you nothing about your municipality?

Because they aggregate situations that have nothing in common. The January 2026 data analysed by the BIPT gives, for 5G: over 95 % of households covered outdoors, over 86 % with a good signal inside buildings, and 69 % with a very good indoor signal, depending on the operator. The three networks combined reach 99.9 % of households.

Those three figures describe the same Belgium. They range from 69 % to 99.9 %. A thirty-point spread that depends neither on your operator nor on your handset, but purely on what someone decided to count.

And averages flatten the extremes. A densely built municipality in Flemish Brabant and a low-density municipality in the Ardennes both enter the calculation, with the same weight per household. The national percentage is the sum of two realities that do not resemble each other.

Where do you find your municipality's 5G coverage rate?

On the BIPT mobile atlas, the Belgian regulator's data portal. It is free, no account needed, and it is the only source that is not a commercial map.

The tool offers two readings. The "By zone" mode accepts a postcode or a municipality name and returns a coverage rate from 0 to 100 %, filterable by provider and by technology. The "Detailed" mode goes down to street level as a grid, showing the expected signal strength cell by cell.

The data comes from the operators, then is corrected by drive tests carried out by the BIPT itself. It is republished twice a year; the survey online as we write dates from January 2026. One detail matters: the atlas is based on the most favourable conditions, with the network unloaded. That is not what you will experience at 6 pm on a Friday in a station.

What do the four signal levels in the atlas mean?

Four tiers, and it is the single most useful reading key in the whole tool.

A municipality can show solid green on "sufficient signal" and empty out almost entirely on "very good signal". That is exactly the gap between the 95 % outdoors and the 69 % very-good-indoors recorded nationally. If you work from home, in a room at the back of the house, it is the last row of the table that decides, not the second.

Territory, population or households: which denominator should you pick?

Households, in almost every case. The atlas lets you choose between three calculation bases, and the gap between them is the first trap in Belgian coverage.

Territory coverage counts square kilometres: it collapses in the Ardennes, where a lot of surface houses few people. Population coverage counts inhabitants. Household coverage counts dwellings — the closest to your real question, which is whether it works where you live.

That is why you still read, on undated .be pages, that "only 65 % of Belgian territory has 5G coverage" next to an article announcing 95 % of households. Both sentences can be true at once. They simply are not talking about the same thing, and the one that concerns you is the second.

Why do two maps give two results for the same address?

Because of a network management technique the BIPT documents itself: "RSRP boosting". Some operators use it to improve the performance of their 4G and 5G networks. Side effect: their coverage appears better on maps than that of operators who do not use it, and the number of bars shown on your smartphone goes up.

The regulator states in black and white that a signal shown as satisfactory does not guarantee a satisfactory experience. That is why it supplements the atlas with campaigns measuring the real-world experience in the field.

The practical consequence is simple: compare the three networks on the BIPT atlas, never on three operator maps placed side by side. The Proximus and Orange maps are useful for a final check at the exact address, but they are not built to be compared with one another.

Which network works best in your municipality in 2026?

It depends on your municipality — but the national gap has closed, and that is new. On January 2026 data published by the BIPT in May, indoor 5G coverage saw what the regulator calls explosive growth: Orange reaches close to 92 % of households, Proximus follows very closely, and Telenet went from 18 % in late 2023 to 90 %. Outdoors the three are near 100 %.

Once the network is settled for your address, price becomes the only criterion left. On the Orange network, hey! Telecom lists 50 GB at €14 a month while Orange Mobile Smart asks €24.99 for 60 GB, prices recorded on 18 August 2026, excluding launch promotions. The signal is the same bit for bit. Our 5G plans ranking lines those prices up at month thirteen, with the host network shown next to each row.

And DIGI, in which municipalities?

Impossible to read in the atlas for now: the BIPT states that DIGI's coverage map is not yet available and will be published shortly. The regulator has not included it in the 5G atlas either, the operator still relying largely on another player's 4G network.

What is measured does appear in the May 2026 report: DIGI meets its coverage obligation by reaching 34.4 % of the Belgian population. In other words, roughly two out of three Belgians who sign with DIGI currently run on national roaming rather than on the operator's own masts. In town, the €9 a month for 60 GB holds up perfectly. In a rural municipality, as long as the map is unpublished, it is a bet.

What can you do if your municipality is poorly covered?

Three steps, in this order, none of which costs money.

First, redo the reading at the right signal level. Many people conclude "no 5G here" while looking at very-good-indoor coverage when a good signal is more than enough for ordinary use from the living room.

Next, test your routes, not only your home. The atlas's detailed mode accepts a street search: run it on your address, on your workplace, and on the two or three places where you spend the most time. 5G on a train has nothing to do with 5G in a living room — between late August and mid-November 2025 the BIPT recorded over 97 % of calls established on sixteen main lines, but speeds appreciably lower than those measured in a car, with more interruptions, the Liège-Welkenraedt and Namur-Arlon lines being the least stable.

Finally, use Belgian law. Since July 2022, cancelling a telecom contract is free after six months, number kept via Easy Switch, and the new operator handles the switch. Six months of a real subscription in your municipality is the one coverage test nobody can argue with — and the only one that includes network load, which the atlas ignores by design.

We do not test plans in the field: we compare public price lists and the measurements published by the regulator. For your street, the atlas decides; for your bill, the comparison table.

Sources: BIPT, mobile atlas (bipt-data.be), January 2026 survey, consulted on 18 August 2026; BIPT, "The 5G network covers more than 95 % of Belgian households depending on the mobile operator", May 2026; BIPT, user experience measurement campaign, late August – mid-November 2025; RTBF/Belga, 7 May 2026; ITdaily, 7 May 2026; operator price lists recorded on 18 August 2026; municipalities of Belgium (Wikipedia).