Skip Navigation

Ofcom thinks fastest speed in UK is 350 Mbps
Monday 08 December 2014 13:13:45 by Andrew Ferguson

Ofcom has published its 2014 Infrastructure Report and a new interactive map. As always with large datasets presentation is always a problem, but there is lots of scope for confusion we believe in the new maps, since the map site is only mapping based on a postcode sector, which is can be an area covering several square miles and we all know how broadband speeds can vary from house to house.

The statistics since they are from Ofcom will be widely used, but the publication is already 6 months out of date and the headline that 75% of UK households can get superfast broadband (78% when you combine FTTC and Cable without the speed qualifier) has been surpassed by the 80% figure released by the DCMS. Of course we don't rely on just the figures from Ofcom or the DCMS and many will dispute that its just 3% that fail to get superfast speeds from FTTC, but running our own numbers we believe 4.4% of those able to get FTTC get speeds below 30 Mbps. The UK broadband landscape is also rapidly changing at a rate of 45,000 extra households getting access to FTTC per week, which fits in with the 40,000 rate from BDUK and a smaller number of commercial cabinets being enabled.

The report does mention Gigabit suppliers like CityFibre, Gigaclear and Hyperoptic but seems to neglect them when talking about the national broadband speeds, since they only list 350 Mbps as the maximum speed when it should be 1 Gbps.

We need to highlight that the speed information is calculated from line data supplied by multiple operators, which in theory should give a much better idea of the situation in rural areas than the SamKnows testing system, though there is still the disjoint between what is delivered to the modem and then the quality of the wireless connection which for millions is their many connection method in the home. The 2014 report has taken a much deeper look into how FTTC performs over distance, something thinkbroadband has been clear about for some years now.


Posted by rtho782 over 2 years ago
On downloading the CSV, it seems to have no data for any BS postcodes.

There are also a grand total of 5 postcodes in the entire UK with more than 152mbit average, which seems wrong - Hyperoptic and Gigaclear alone cover more than 5 postcodes!!

Posted by rtho782 over 2 years ago
Sorry, that's 4 postcodes.

Also, over the whole country, there are a grand total of 209 postcodes with a max speed of 300mbit. Nobody on 330mbit. I'm convinced this data is just plan wrong.
Posted by Dixinormous over 2 years ago
At least some postcodes with 152Mb average will be trials. From anecdotal reports hardly anyone in the 'commercial' Openreach FTTP areas purchased 300Mb and few 200Mb. Most settled for 80Mb or 40Mb.

I imagine the same will go for BDUK FTTP.

As far as Hyperoptic/Gigaclear go other packages than 1Gb are available and people in UK don't like paying for broadband.
Posted by TheEulerID over 2 years ago
It's a bit difficult to see why a domestic customer would currently buy higher priced 1Gbps packages at the moment apart from boasting rights (ok that's important to some) as there aren't many services out there which will get anything near that rate.
I suppose there are torrent type services, but I wonder what real world data rates are rather than running benchmarks.
Posted by Dixinormous over 2 years ago
Speaking to a couple of people download speeds upwards of 500Mb/s are not uncommon on well-seeded torrents.
Posted by rtho782 over 2 years ago
I'd buy a 1gbit service.

Gigler do it for £25 a month (their top 1tb package) in Bournemouth.

Hyperoptic charge £50 for 1gig or £25 for 100m, I'd take the gig there as well, 2x the price for 10x the speed, and would make things like downloading from Steam much faster. I also wouldn't have to pause Steam to watch Netflix 4k, which recommends 25mbit alone, etc.

BT charge £50 a month for their 330m service. I'd take that over £40 for the 200m.

Plenty of people going for VM say it's because 152 is double 76, there is certainly demand for a speed over 100m.
Posted by mdar5 over 2 years ago
Some if not most of the Oxfordshire post codes appear missing.
Probably explains the absence of a Gigaclear 1Gbs connection.........

I can only find OX 12,17,15,27,33,39,52 and 93.
Posted by George99 over 2 years ago
The data is findermentally flawed.
It should be based an addresses not postcodes.....then we will see the swathes of addresses that are STILL stuck with 1-3Mbs....even in London
Posted by gerarda over 2 years ago
Ofcom data is always flawed. Just another report to be filed under P for Propaganda
Posted by galacticz00 over 2 years ago
My post code doesn't appear on the spreadsheet. Perhaps this is because SFB=0, 4G=0, 3G=2 providers but 0 data, full freeview=0, digital radio=0
Posted by Spud2003 over 2 years ago

The big ISPs like BT and TalkTalk apparently don't just see their future as bog standard ISPs, they want to be like cable providers and deliver content and other services. It would help them if they had high capacity networks, so although I sort of agree that most people technically don't require 1Gbps connections today I think the future commercial requirements of ISPs are heading in that direction at least.

If in the future, for instance, people could buy a completely 4K TV package with a 1Gb connection instead of shelling out for Sky I could see them paying a bit more.
Posted by Blackmamba over 2 years ago
Hi Broadband Watchers
The data and write up is very fair but it is out of date by nine months on the fixed line broadband.
Posted by gerarda over 2 years ago
The data and write up is the opposite of fair.
Posted by WWWombat over 2 years ago
People seem wary of paying for super-mega-brilliant-def TV content. They might have bought HD televisions, but they're not buying the HD content.

Look at sales figures for DVD vs BluRay. Last figures show DVD is somewhere around 7-8x BluRay sales.

That 4K TV package will probably work perfectly well within a 100Mbps package, with room to spare, but it'll be a long time before a reasonable percentage of the population would even bother.
Posted by timandhaylea over 2 years ago
Loving the BBC News story on this (

"If the length of fibre cable from the household to the nearest junction cabinet is too long, download speeds of 30Mbps are not possible."

Glad to see they put some effort into getting their facts right before publishing!
Posted by RuralChris over 2 years ago
The Postcode level data supplied by OFCOM is missing data from the whole of the South West and large parts of Yorkshire. However data is supplied on the Local Authority spreadsheet for the 'missing' areas.

In the South West no postcode data is supplied for BA (Bath), BS (Bristol), DT (Dorchester), EX (Exeter), GL (Gloucester), PL (Plymouth), TA (Taunton) and TQ (Torquay).

In Yorkshire and Humber Region 8 of the 11 Postcodes are missing; DL (Darlington), HD (Huddersfield), HG (Harrogate), HU (Hull), HX (Halifax), LS (Leeds), WF (Wakefield) and YO (York)
Posted by andrew (Favicon staff member) over 2 years ago
Apparently Ofcom has said there was a problem with the CSV and it will be updated.
Posted by mdar5 over 2 years ago
Thank you Andrew for taking the time to chase that one up with OFCOM
Posted by gerarda over 2 years ago
Ofcom just seem to make up the numbers and don't even use their own data. The report quotes the percentage of sub 2mb lines as 3% but the local authority spreadsheet gives a figure of 1.23m out of 20.9m which is 6% or twice their quoted figure.

If you add in all premises using satellite, FWA, and 3G because they cannot get a 2mb line then the real figure is probably nearer 10%
Posted by Blackmamba over 2 years ago
Hi Gerarda.
I feel the 3% figure is a true result as it covers all of GB I will quote again that a customer on an exchange which is 21CN enabled the chances that they will be unable to get -2% is very remote,just check Thinkbroadband Maps.
Posted by gerarda over 2 years ago
Blackmamba- just download the local authority spreadsheet and add it up for yourself

Posted by WWWombat over 2 years ago
You're comparing apples & oranges...

The LA figures that add up to 1.23m are take-up figures, not availability. The 3% figure is availability.

To make the percentage, you need to divide by the no. of properties in the UK, not by the no. of broadband lines. That turns your 6% into 4%.

In the report, the 4% is shown in figures 16 & 17. There it also shows that, of the 4%, 2% could take superfast if they chose to, and 2% don't have that option.

Likewise can't assume all satellite, FWA and 3G users have only sub-2Mbps lines.
Posted by WWWombat over 2 years ago
Figure 20 breaks down VM's broadband customers:
10% take the 120 or 152 packages
27% take the 60 or 100 packages
63% take the 20, 30 or 50 packages

Same figure shows the FTTC breakdown:
32% take (and get) packages above the 40Mbps level
48% take (and get) packages at the 40Mbps level
20% take packages at the 40Mbps level, but get less

In both camps, it seems that two-thirds of people only want to pay for the entry level packages.
Posted by gerarda over 2 years ago
No you are comparing apples and oranges. Of the 20M properties surveyed by Ofcom 6% are sub 2mb. To get to 4% you have to assume that all the other 10 million could all get above 2mb which is nonsense.

I am not assuming that all satellite, FWA and 3G users have only sub-2Mbps lines just that ofocm have excluded these and they need to be added to the numbers where they would otherwise not be able to get 2mbps
Posted by gerarda over 2 years ago
if you take my postcode as an example there are 6 premises, ofcom show 2 sub 2mb (actually sub 0.6mn) there are 2 using 3G, one using FWA and one without broadband. None of the other lines can support an ADSL connection. So Ofcom has understated the sub 2mb lines by two thirds

Posted by WWWombat over 2 years ago
Nonsense indeed.

Your original issue was that Ofcom just make up the numbers, and don't use their own data.

I'm showing that they do use their own data, and are consistent in applying it. The figures with 4% sub-2mbps fixed-connection properties also shows the non-fixed-connection properties elsewhere in the figure. 4% is the right figure to use there, and matches the 1.23m you added up in the spreadsheet. All these figures relate to take-up.

Right numbers, used in the right way, with consistent comparisons.
Posted by WWWombat over 2 years ago
You choose to use a different denominator, then compare take-up vs availability. Wrong numbers, used inconsistently.

Unfortunately, you are so obsessed with calling out Ofcom's bad statistics, that you are making worse mistakes than they are.

Your postcode example is more of the same. Looks to me like Ofcom correctly counted the properties with take-up - leaving 4 out that have no fixed connection.

At that point, it isn't counting availability of sub-2Mbps via fixed connection, or even availability of sub-2Mbps via any connection type. It hasn't understated anything.
Posted by gerarda over 2 years ago
@wwwombat. I just do not see where you are coming from.
Ofcoms figures show 1.23m sub 2mb based on a survey of 20.9m. If they use a figure of higher than 20.9m they do not have the data on which to do so. You cannot extrapolate the rest.

Applying their logic from the other end says 19.7m lines can get more than 2mbs so only 2/3 of total premises can get more that 2mb. Would you agree with that?

I can find no reference in either the data or the report to non fixed lines. Could you please point me to where you found it?

Posted by gerarda over 2 years ago
Ofcom and BT in their attendance at Parliament last week are using a stat to try to say that any residual sub 2mb lines can be infilled with satellite. in our postcode code example there are 6 premises which need to be upgraded not 2, Whatever way you look at the number of premises that cannot get a line with more than 2mb is higher than 1.23m not lower.
You must be logged in to post comments. Click here to login.