A new report from Thinkbroadband has provided a useful independent assessment of how big – in terms of network size (coverage footprint) – the top 17 largest Fibre-to-the-Premises (FTTP) broadband ISP networks in the UK have become in Q2 2023, which we’ve compared against the official coverage claims below.
Regular readers will already be aware that one of the difficulties with keeping tabs on the rapid UK growth of Fibre-to-the-Premises (FTTP) based broadband ISP networks is that some operators often make coverage claims that are difficult to independently verify. This is relevant because, over the years, we’ve sometimes found that the official “premises passed” figures put out by some providers may not entirely reflect reality.
For example, in some cases those figures could include part-build networks that aren’t fully live yet and, in other cases, the network may be technically built, but customers in some of the covered areas won’t be able to get it ordered or installed by an ISP (i.e. not yet truly “Ready for Service” – RFS). Similarly, providers sometimes make mistakes in their data (example).
Advertisement
The latest quarterly State of Broadband Report Q2 2023 from TBB is thus extremely useful because we get a fairly up-to-date run down of how much coverage has been independently verified to exist (RFS) across the largest alternative network (AltNet) operators in the full fibre space (the list reflects those with over 100k+ premises passed). This enables us to compare TBB’s latest data below with the official coverage claims from the operators.
Naturally, there are a few caveats to this data, which need to be considered for the proper context. Firstly, TBB is not perfect, and they do sometimes miss bits and pieces of network coverage (please email them if you spot this). Secondly, conducting independent analysis of network builds like this is slow and laborious work, thus over the years we’ve tended to perceive that TBB’s latest data can be 2-3 months behind actual build.
Such a time lag, which is likely to vary between operators, may not seem like much, but it can create disproportionately large gaps between independent and official figures. Big gaps are most likely to occur during the early ramp-up period of a new network build, where smaller networks may go from having a few tens of thousands of premises passed to hundreds of thousands, all within the space of a year (a fair few AltNets are still in this phase).
In short, it’s wise not to make the mistake of automatically inferring that a big gap is because an AltNet may be overstating their coverage. In addition, we’ve also dated the official claims below as most operators only very occasionally provide an update on their build progress, which will thus be out of step with TBB’s latest modelling.
Advertisement
Top 17 Largest UK Full Fibre Networks by Coverage – Q2 2023
Operator | Premises Q2 (Q1) – TBB Analysis |
Official Claim |
Openreach (BT) | 10.9 million (9.5m) | 11m – Jul 2023 |
CityFibre | 2.3 million (2.2m) | 2.9m Built (2.5m RFS) – Jul 2023 |
Virgin Media / nexfibre | 1.5 million (1.3m) | 3m (VMO2 + Nexfibre) – Jul 2023 |
CommunityFibre | 1.1 million (858,000) | 1.2m – July 2023 |
Hyperoptic | 1 million (911,000) | 1.15m – Feb 2023 |
Netomnia (YouFibre) | 556,000 (359,000) | 500,000 – Jul 2023 |
Gigaclear | 343,000 (311,000) | 450,000 – Jun 2023 |
Trooli | 283,000 (212,000) | 219,000 – Dec 2022 |
KCOM | 267,000 (265,000) | 300,000 – May 2022 |
G.Network | 255,000 (253,000) | 400,000 – May 2022 |
Fibrus | 238,000 (no Q1 data) | 256,000 – Mar 2023 (240k RFS) |
Brsk | 190,000 (no Q1 data) | 200,000 – May 2023 |
ITS Technology (mostly biz fibre) | 185,000 (185,000) | No recent official data |
Zzoomm | 150,000 (no Q1 data) | 150,000 – Jun 2023 |
Jurassic Fibre | 127,000 (no Q1 data) | 150,000 – Mar 2023 |
Swish Fibre | 127,000 (no Q1 data) | No official data |
OFNL | 126,000 (no Q1 data) | No official data |
We aren’t going to micro analyse each operator above, but most of TBB’s real-world focused estimates of Ready for Service (RFS) style coverage are roughly where we’d expect them to be when compared with official claims. On the other hand, KCOM does seem to be trailing a bit more than expected and that awkwardly large gap still exists on G.Network.
Readers may also notice the huge gap on Virgin Media’s (VMO2) and nexfibre’s combined network. The catch here is that Virgin’s total includes Hybrid Fibre Coax premises that have recently been upgraded to the latest FTTP / XGS-PON connectivity, while TBB is largely only tracking newly built FTTP.
Finally, it’s important to remember that a lot of these FTTP networks are overbuilding rivals in different parts of the country, particularly around dense urban locations. But if we were to just look at the UK coverage of alternative networks (excluding Openreach, Virgin Media and KCOM) then they’d reach just over 24% of premises.
Advertisement
Wow, Community Fibre and Hyperoptic are neck and neck. Not sure the level of overlap, but surely their combined coverage is 60% of London?
No because HO does not cover SDU and HOs London footprint is only around half in London. There is big overlap in the big landlords
There is a big gap in the HO number. Their reported number was feb, not Q2
City Fibre also has too big and suspicious of a gap. I would reckon considering their layoffs to be closer to the TBB number than the reported number
Netomnias number is actually below the TBB number so TBB shouldn’t be that far behind
Hyperoptic do cover SDU, albeit mostly new builds but still
Thanks guys for the clarification
Hyperoptic have covered large areas of Highbury, which are houses and probably a hundred years away from being new build. Lots of effort has gone in but the demand seems to be there for it.
Example of gap cityfibre tells ofcom its available in mdu and when people try and order are told no. With the footprint size this is a large part of the gap. Not the only provider where we see this when trying to figure out gaps.
That’s insane, so they are basically adding all the MDU in their towns towards their numbers regardless of them being done or not
The report predicts 100% fttp coverage by July 2026. Surely not?
Where do you see that on the report ?
top right of page 3
Can happen if pace is maintained
In the real-world, it’s necessary to reflect that builds tend to accelerate, before reaching a peak and then slowing down over the remaining years. So I think Andrew is following a linear curve based on current progress, which doesn’t weight/predict for the ramp-down side, as that’s hard to estimate. Please correct me if wrong, Andrew.
I’d say it’ll take a few years longer than 2026 to get close to 99% (I’m not saying 100% as we already know that a small proportion of properties will always be missed for various reasons).
There will be a curve and the projection is linear, but if things slow down then this will feed in, and the 2026 will slip to a later date.
Need to watch apples with apples here. Particularly providers like ITS, focused on Business. Having been one of their partners, they talk up a good story about premises passed (like their CEO blags a lot of things), but reality is mainly Ethernet provision not FTTP from distant DPs, so premises are not that well served and most often come with a No or high ECCs when you place orders. Partner satisfaction with order delivery is pretty low.
Last week I sat in on a Gigaclear webinar where their evangelist announced latest plans for the rollout of FTTP in my local town. I did them the courtesy of not replaying portions of the screen recording (Camtasia) from their previous webinar circa 18 months ago. The slippage would have been embarrassing to my hosts.
I was amused to hear their two experts proclaim that 80mbps was the max speed over copper wire. To help them out I mentioned I was listening over GFAST. “Oh what’s that…?”
The Webinar was useful as it triggered my curiosity. I checked out their international phone call tariff. Ouch!
OFNL have a much larger footprint as I understand, more like 500k RFS
If there is an easy 300,000 premises very happy to add them, get me a few postcodes of the missing ones.
I wonder how accurate the Trooli figures are – I’ve sent a few emails to Andrew about this – They show large parts of my town and other near by towns as covered but you dig a little deeper and the number of skipped streets is concerning. I know this as they declared my street covered to Thinkbroadband but it was unavailable to order. Andrew at TB removed it. I know its not just me as I have a colleague in a nearby town who’s house was skipped because they seemingly ran out of connections at the end of his road, so just left off his and 4 other properties. I’m aware there is a real desire to inflate these numbers for investors but ultimately these altnets will happily fudge the numbers to suit themselves.
Who’s Who . . . in the UK Broadband provider “Biggest fibber” stakes.