Skip Navigation

BDUK publishes updated take-up figures for superfast broadband projects
Wednesday 08 June 2016 17:39:14 by Andrew Ferguson

The claw back system of the various BDUK projects looks set to release even more money over time if the take-up rates reported by the BDUK team for the projects across the UK continue and given the volume of moans from those who don't have access to a superfast service yet, the suggestion is that any cabinets going live now may see even more rapid take-up than ones installed back in 2013/2014. The difference in 2016 is the amount of online TV viewing driven by changes such as BBC Three moving to be online only and the vast catch-up libraries.

ISPreview has been keeping a tally of the reported take-up rates from the BDUK projects for a while and with the release of the March 2016 take-up figures by the BDUK team today we thought we would see what we deliver in terms of verification of the figures reported. Our observations on take-up are driven by analysing speed tests people have run in a quarter, which while not perfect should provide a good observation mechanism and if there are massive disparities we will then investigate areas more closely (for example we may need to extend the analysed period to 6 months).

BDUK Project Area - Phase 1 Projects BDUK Reported Uptake March 2016 thinkbroadband Observed Uptake Q1/2016 thinkbroadband Mean VDSL/FTTC Download Speed Q1/2016
Berkshire 31.5% 33.5% 28 Mbps
Bucks and Herts 29.9% 32.4% 28.4 Mbps
Cambridgehire & Peteborough 35.2% 31.9% 27.5 Mbps
Central Beds, Bedford and Milton Keynes 29% 30.7% 26.9 Mbps
Cheshire 32.6% 28.6% 28 Mbps
Connecting Devon and Somerset 22.5% 23.3% 27.8 Mbps
Coventry, Solihull & Warwickshire 32% 29.6% 28.9 Mbps
Cumbria 25% 28.4% 27.6 Mbps
Derbyshire 20.1% 27.6% 26 Mbps
Dorset 21.9% 26.9% 27.5 Mbps
County Durham 21% 28.2% 26.3 Mbps
East Riding of Yorkshire 27.2% 28% 27.2 Mbps
East Sussex 27.6% 27.8% 28.7 Mbps
Essex 23.6% 27.6% 26.9 Mbps
Greater Manchester 24.7% 28.9% 28.4 Mbps
Hampshire 28.2% 27.5% 28.6 Mbps
Herefordshire & Gloucestershire 22.9% 22.2% 28.4 Mbps
Highlands & Islands 21.6% 21.4% 27.2 Mbps
Isle of Wight 19.8% 29.4% 26.3 Mbps
Kent 26.7% 27.7% 27.8 Mbps
Lancashire 31.1% 30.5% 28.3 Mbps
Leicestershire 27.1% 30.4% 27.7 Mbps
Lincolnshire 27% 27.4% 26.5 Mbps
Merseyside 24.4% 25% 27.4 Mbps
Newcastle Upon Tyne 25.7% 28.2% 28.2 Mbps
Norfolk 28% 28.7% 27.8 Mbps
North Lincs & North East Lincs 30.6% 32.6% 27.5 Mbps
North Yorkshire 32% 28.4% 28.6 Mbps
Northamptonshire 36.7% 35% 29 Mbps
Northern Ireland 19.3% 21% 25.9 Mbps
Northumberland 28% 28.3% 29.5 Mbps
Nottinghamshire 26.8% 29.2% 26.1 Mbps
Oxfordshire 26.1% 32% 30.4 Mbps
Rest of Scotland 20.9% 22.9% 27.7 Mbps
Rutland 51.7% 38.5% 29.6 Mbps
Shropshire 27% 23.5% 28.6 Mbps
Staffordshire 23.9% 26.8% 27.9 Mbps
Suffolk 31.5% 26.6% 27 Mbps
Surrey 42.2% 34.4% 29.4 Mbps
Wales 22.4% 26.6% 28.2 Mbps
West Sussex 23.2% 29.9% 29.5 Mbps
West Yorkshire 27.7% 24.7% 26.4 Mbps
Wiltshire, South Gloucestershire 37.5% 29.5% 27.8 Mbps
Worcestershire 25.6% 24.7% 27.6 Mbps

As with any data analysis there is always factors to take into account, e.g. for the BDUK figures take-up can fluctuate depending on whether a project has just delivered lots of new cabinets. For the thinkbroadband observations there is always the possibility that only those with a fast connection have tested their connection or only those with problems are using the speed test, plus many people may only run a speed test when switching provider. One variable that we will work on over the next few months is identifying which chunks of FTTP have been delivered using commercial money and those with BDUK funding, given the low volumes of FTTP delivered this should not make a massive difference to the figures.

The degree of correlation between the official and our observations is encouraging and adds some more data into the debate on whether the BDUK projects are delivering anything and do they represent value for money.


Posted by TheEulerID 11 months ago
I've just put the BDUK & TBB observed takeup figures into a spreadsheet and put a linear regression line into the two sets. There is a positive correlation, but it's not that robust (pearson coeeficient of about 0.54 and with a gradient of just 0.43 (when it ought to be about 1). Also, it shows an intercept at 16.5% with BDUK penetration of 0%. That's clearly not possible. My conclusion is that the TBB figures are a good indication of direction but not of absolute values, very possibly down to the self-selecting sampling set.
Posted by andrew (Favicon staff member) 11 months ago
So a waste of time and throw into the bin? Or a useful adjunct to show whether official data is being made up from thin air or might actually be true?

As with coverage, exercise was to see if the strongly held assertions that the BDUK process is a failure and not delivering are closer to the mark than the official announcements.
Posted by gerarda 11 months ago
If you are referring to assertions made by the likes of me, the BDUK process failed when it made the decision to leave 5% in a relative digital dark ages. Coverage stats for the other 95% are irrelevant to that view.
Posted by TheEulerID 11 months ago
No, not a waste of time, which is surely not a reasonable interpretation of what I wrote. It clearly shows the same direction and progress using an independent measure. I just mean it can't be used to check absolute numbers.

As far as gerarda's point is concerned, then this is obviously not addressing the final 5% (or will it be 3%) issue, but that's better dealt with under USO & coverage issues, not this item.
Posted by Michael_Chare 11 months ago
Is there anything which shows what performance increase makes people more likely to adopt FTTC?

Also it would be nice to see some trend data. Does the demand increase substantially reduce when you get to 30% or is the overall demand likely to reach say 40%?
Posted by andrew (Favicon staff member) 11 months ago
@TheEulerID Was more asking so you could expand on what you said initially, since in current cliemt 'not that robust' means many will assume its to be given an ignore tag.
Posted by andrew (Favicon staff member) 11 months ago
Who knows if take-up gets to 50%, we might see the 95% - maybe to be 97% turn into 98%. Or at the end of the 7 year contracts councils to actually be given cashback.
Posted by Blackmamba 11 months ago
Hi Broadband Watchers.
As TBB results are just a sample size and are an average over 3/6 month on post codes they I feel are showing low results which is a good and fair result for Openreach. As they have updated their data collection future results will close this gap as more tests are made , they are also highlighting the under 10 meg locations which are in the step two programs .
Posted by andrew (Favicon staff member) 11 months ago
@blackmamba I'll forego the usual mistakes and ask What is the step two program?
Posted by TheEulerID 11 months ago
It should (presumably) be possible to analyse the data for step increases provided it's been recorded against a username on the site. It would provide an interesting stat to show what sort of range of increase is being seen. After, of course, the data has been suitably anonymised.

Posted by andrew (Favicon staff member) 11 months ago
Registered users are a self selecting sample of a self selecting sample of a self selecting sample. Plus we operate speed tests for a growing number of other sites which have no registration system.
Posted by Blackmamba 11 months ago
Hi Andrews Staff.
I would class the step Two i(phase 2) in Surrey as the existing money in the OMR pot plus the clawback which will be used to upgrade all lines (500k) to get over 15 meg if possible as long as the money holds up so they hit the target of 99.7 % (15 meg) as quoted at Dorking meeting.
Posted by CarlThomas 11 months ago
You're thinking of steps 3, 4 and 5, BM.
Posted by Blackmamba 11 months ago
Hi Carl.
I did state Surrey area which has two stages and also their deployment was implemented early and was working to a different workload but was over all controlled but the EU regulations.
Posted by CarlThomas 11 months ago
Even though you didn't mention it at all I gathered that. You incessantly bang on about Surrey whenever anything even remotely related to BDUK is mentioned here.

Think most readers are more familiar with Surrey's BDUK programme, or at least your view of it, than their own.

At least you have noticed now that Surrey is working differently to other areas, so hopefully aren't going to keep extrapolating it to everywhere else. Surrey is a very different animal from Scotland, East Yorkshire, etc.
Posted by jt2354 11 months ago
West Lothian is not on the list. Scotland divided into Highlands and islands and "Rest of..."?
Posted by andrew (Favicon staff member) 11 months ago
That is because the two different BDUK projects in Scotland are HIE and Rest Of Scotland

More detail as ever at but no take-up splits. Need to see how the data performs over a another quarter or two.
Posted by Blackmamba 11 months ago
Hi Andrews Staff.
I have just noticed that today's results have been calculated to using two decemial points and it has dropped the under 2 meg by a high % plus giving a better perspective across the country's cabs results.
Posted by silverdove 11 months ago
No Cornwall? No surprise there then!

Just checked my postcode in the hope that Superfast will be here within the next 10 years and got :
"It looks like superfast broadband may not be available in your area yet, but it could be coming to you soon through Government and local authority investment."

Ha Ha - BT just sent me an email "We know you want broadband that moves as fast as you do, so we're upgrading your BT Infinity to speeds of up to 52Mb".

Might take them up on that as my current speed is 1.65mb :-))

What a joke!
You must be logged in to post comments. Click here to login.