Broadband News

We reveal size of proposed BT USO footprint

The latest BT proposals on the Universal Service Obligation have led to Ofcom starting a consultation on the price impact of that proposal to help inform the eventual decision of whether to accept this proposal or hold out for something different.

We have been tracking the number of premises under 10 Mbps for a long time and it has been gradually reducing e.g. since May 2016 we have seen it fall from 1.3 million premises to 888,900 premises but the BT USO proposal has specified a minimum 1 Mbps upload speed which potentially means ADSL2+ services even if offering a download speed above 10 Mbps would not meet the standard. ADSL2+ can run at upload speeds just above 1 Mbps but generally this is just the longest lines and the reality in actual use once overheads are accounted for is that people do not see above 1 Mbps, ADSL2+ Annex M can change this but would only be relevant for lines with download speeds above 16 Mbps.

We hinted on Thursday an analysis of what excluding ADSL2+ from the data set would mean, particularly after we raised concerns over the projection Ofcom was using, i.e. 90% of those not getting superfast were likely to need USO work. We now can publish the data for the UK and its nations and regions

thinkbroadband calculation of superfast coverage and premises that currently fall under proposed BT USO
Based on coverage levels from 9th August 2017
Area% superfast
24 Mbps or faster
% Under 10 Mbps DownloadUnder Proposed BT USO
10 Mbps down with 1 Mbps upload
North East 96.3% 1.3% 2.7%
927,552 Premises 893,634 12,094 25,419
London 96.1% 0.8% 3.2%
4,397,705 Premises 4,224,476 36,791 139,745
East Midlands 95.9% 1.9% 2.7%
1,101,327 Premises 1,055,946 20,570 29,485
South East 95.7% 1.6% 2.7%
2,154,787 Premises 2,063,109 34,810 57,381
West Midlands 95.5% 1.9% 3.2%
2,767,347 Premises 2,641,699 52,155 87,889
North West 94.8% 2.6% 3.7%
3,895,748 Premises 3,693,202 100,089 142,475
England 94.1% 2.5% 4.3%
23,891,718 Premises 22,475,232 601,127 1,032,516
United Kingdom 93.4% 3.1% 4.9%
28,515,548 Premises 26,623,335 887,324 1,387,154
East of England 92.3% 3.5% 5.6%
2,602,516 Premises 2,401,172 90,202 145,791
Wales 91.5% 4.9% 6%
1,321,435 Premises 1,209,092 64,740 79,142
Yorkshire and Humber 91.5% 3.9% 7.1%
2,557,452 Premises 2,341,191 99,727 180,802
Scotland 90.8% 5.3% 7.4%
2,575,929 Premises 2,339,059 137,014 191,053
South West 90.6% 4.4% 6.4%
3,487,284 Premises 3,160,803 154,689 223,529
Northern Ireland 82.6% 11.6% 11.6%
726,466 Premises 599,952 84,443 84,513

Ofcom is suggesting that by 2021 the number needing USO intervention would be around 785,000 premises (based on a premises figure of 29.09m which includes estimated premise growth over the next few years) based on the theory that as 90% of those not having a superfast broadband option in May 2016 were below the target that once the roll-outs which 97% superfast coverage you can use the same 90% reducer. While writing up the Ofcom price analysis we thought that the change in the pattern of where and how superfast broadband is being deployed i.e. more FTTP and infill VDSL2 cabinets than previously that this model would not hold and our analysis based on the coverage we knew about on 9th August suggests that this 90% has already changed to 73% and we believe it may improve further

Being aware that the number needing USO work may well be lower is important as if less people need help the proposals to spend some £450m to £600m can be utilised to do things like deliver more full fibre and most importantly reduce the number of premises that might still end up relying on satellite broadband to below the suggested 0.3% of UK premises.

We will report publically as to how the premises counts falling within the proposed BT USO specification is changing and with a number of weeks before the consultation submission date we may be able to show that even over a period of a few weeks the simplistic model Ofcom used is becoming less accurate.

The regional analysis shows that the change is not uniform, so the areas such as Southwark with a high level of exchange only ADSL2+ lines add to the London picture. The high levels of VDSL2 coverage in Northern Ireland have a different effect i.e. it is so widespread now that there is no real detectable difference from our analysis, or expressed another way VDSL2 while not delivering superfast speeds to lots of people in Northern Ireland is offering many something that is at least better than ADSL or ADSL2+.


Yeah lets screw those people on less than 10 Mbps and use that money to fibre up premises that already have access to superfast speeds.

  • galacticz00
  • 11 months ago

Thanks, It will be good to see your forecasts if Scotlands R100 is assessed, the Welsh plan is also comprehensive and the £150m for NI should also reduce the costs associated with BT is proposing.

  • ValueforMoney
  • 11 months ago

Another 3 years to wait just to be told to piss off and get satellite!

  • nobroadband
  • 11 months ago

Andrew, You numbers are significantly less than the Governments and Ofcoms where by the end of 2016 10/1 scenario was 2.6m. They are modelling off a premise count of 30+m which I think creates the difference and why I was posing the question earlier in the year.

Something as simple as an incorrect premise count will would have a tendency to reduce the reduce the ambition.

  • ValueforMoney
  • 11 months ago

Post a comment

Login Register