» ISP News » 

UPD Rural vs Urban Gap Closing as Average Broadband Speeds Hit 18.7Mbps

Friday, October 3rd, 2014 (10:52 am) - Score 2,940

Ofcom’s latest biannual report into the real-world speeds of fixed line broadband ISP consumers across the United Kingdom reveals that the Internet performance gap between rural and urban areas is still wide, yet their speeds are now improving at a faster rate (+20% rural vs +5% urban / suburban). Overall the average download rate has risen by just 5% to reach 18.7Mbps (up from 17.8Mbps in the last report).

It’s important to stress that the telecom regulators report reflects data gathered to May 2014 and is arguably more accurate than most because it uses real-world information collected via custom SamKnows monitoring routers, which have been installed in homes across the country. But this must be weighed against the fact that only 2,175 homes have installed such a device, which altogether conducted 2.7 billion separate tests.

The report also includes experimental analysis of Internet connection reliability (service disruptions and disconnections), which found that older style ADSL broadband lines experienced an average of 0.5 disconnections lasting longer than 30 seconds per day. By comparison customers on “superfast” (30Mbps+) fibre and cable broadband packages experienced less disruption, an average of 0.1 disconnections lasting longer than 30 seconds per day. Ofcom intends to expand upon this in future reports.

Otherwise it’s no surprise to find that broadband speeds, including the average upload rate that increased to 2.4Mbps (up by just 5% from 2.3Mbps at the last report), have continued to rise. The ever faster speeds being delivered over Virgin Media’s 152Mbps capable cable (DOCSIS) network and the growing availability of BT’s up to 40-80Mbps FTTC lines continue to drive the majority of increases, while older 8-20Mbps ADSL / ADSL2+ services remain fairly flat. But the impact of crosstalk interference might be hurting FTTC, which has seen its average performance (dominates FTTx below) fall to be overtaken by cable.

ofcom average uk broadband isp speeds by technology q2 2014

Meanwhile rural areas are continuing to lag behind their urban counterparts, which is to be expected. The Digital Speed Divide exists because so far most of the broadband performance improvements, even those funded by the state aid fuelled Broadband Delivery UK (BDUK) scheme, have predominantly focused on urban and sub-urban areas.

Thankfully that is starting to change and so we are now seeing a more noticeable improvement in rural areas as older ADSL lines are replaced by FTTC and some ultrafast FTTP (BDUK aims to bring 24Mbps+ fixed line superfast broadband to 95% of the UK by 2017). But in terms of raw speed there’s still a big gap to be bridged.

ofcom average uk rural vs urban broadband speeds q2 2014

According to Ofcom’s analysis, rural customers experienced a bigger increase in average broadband speeds (up by 20% to 13.6Mbps) than people in urban and suburban areas (up by 5% to 33.4Mbps and 22.9Mbps respectively).

Specific ISP Performance and Latency

As usual Ofcom’s report, being based on such a small sample size, can only reflect the performance of the largest broadband ISPs, which is an important consideration because smaller ISPs tend to pay more attention to quality and would probably do well; especially in the next few tables that track individual ISP performance by connection technology.

ofcom average download speed by isp q2 2014

As usual there’s not a lot to choose between the ADSL2+ based providers, which all exhibit fairly similar performance because it’s a saturated market and most are already running up against the limits of their existing copper lines, while the superior top speeds of Virgin Media’s cable network clearly lead the big boys corner in terms of download speed.

However the fastest FTTC lines do tend to have better upload speeds (not shown above) of up to 20Mbps, which compares with 12Mbps on Virgin’s 152Mbps option. Meanwhile both Virgin’s cable and the FTTC packages on BT’s infrastructure seem to deliver average speeds that are fairly close or above their respective packages headline (advertised) rate. By comparison the 9Mbps or so from older and less reliable ADSL2+ services remain well below the services theoretical top speed of 20Mbps+.

The next area of performance we’ll look at concerns Latency, which is a crucial measure of the time (delay in milliseconds / 1000ms = 1 second) that it takes for a single packet of data to travel from your computer to a remote server and then back again (ping). A lower score is always best (shortest time) and this measure is especially important for fans of online multiplayer games, where a low ping gives you smoother and more accurate gameplay.


The good news is that there’s not much to choose between the ISPs here and even older ADSL2+ services deliver reasonable latency (ping) performance. In fact if this chart included an ancient 128Kbps ISDN line then you’d probably still see many of the listed ISPs above being beaten by it because this is one area where pumping out lots of Mbps isn’t as crucial as the efficiency of the underlying digital network.

True fibre optic (FTTH/P) networks also tend to work very efficiently for low latency, although sadly Ofcom’s report doesn’t highlight these connections because none of the big ISPs have a significantly large enough pure fibre optic network (i.e. not enough data for useful statistics).

Speed Complaints

Unfortunately not everybody gets the speed their ISP promises to deliver and thus it’s important to be aware of the rules that could help. Firstly, the Advertising Standards Authority’s guidelines for tackling misleading promotions of “up to” broadband speeds (here) require ISPs to “demonstrate that [their] advertised speeds are achievable by at least 10% of users“. This is why so many providers now promote a “typical” speed (average performance based on that 10%) instead of the technologies theoretical top speed.

At the same time Ofcom’s 2010 Voluntary Broadband Speeds Code of Practice (Version 2) requires member ISPs (Listed Here) to explain to new customers the access line speed that they’re likely to achieve at home, and to try to resolve any problems when speeds fall significantly below the estimate. If the ISP fails to resolve the problem then customers should be able to leave their ISP, without penalty, within the first 3 months of a new contract. But switching ISP without fixing the underlying issue first may not result in better performance, although a new provider may still show more willingness to help.

On top of that the regulator has recently imposed new Quality of Service standards upon BTOpenreach (i.e. faster installs, quicker repairs and making available more information), although the new measures typically focus more on tackling connectivity problems than end-user performance woes (here). Ofcom are also making it easier to switch ISP, but this system won’t be ready until June 2015 and then ISPs will probably be given another year to introduce it (here).

Generally the rules are far from perfect and Ofcom are currently making some revisions to their speed code (here), although we won’t know the outcome of that until later this year (it’s not expected to be a significant change). In the meantime it’s also worth remembering that consumer broadband is very much a “Best Efforts” service, where capacity needs to be shared between many users in order to be affordable. Customers that want something better still have to look towards expensive business leased lines, which aren’t really practical for domestic needs.

Ofcom’s UK Fixed Line Broadband Speed Study (May 2014 Data)

UPDATE 12:11pm

A quick comment from the Government and Virgin Media.

Ed Vaizey, Digital Economy Minister, told ISPreview.co.uk:

I’m delighted to say that these findings, whilst fantastic news, are not unexpected. Government has taken access to superfast broadband speeds to more than 1 million homes and businesses across the UK through our nationwide rollout, which has been instrumental in average speeds increasing by more than 3 and a half times since May 2010.”

Tom Mockridge, Virgin Media’s CEO, added:

We’re delighted to be officially rated the UK’s fastest broadband provider six years in a row. When it comes to making the most of our increasingly connected lives, speed matters. Virgin Media continues to invest in bringing superfast broadband to Britain at home, at work and on the go.”

Share with Twitter
Share with Linkedin
Share with Facebook
Share with Reddit
Share with Pinterest
Mark Jackson
By Mark Jackson
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on Twitter, , Facebook and Linkedin.
Leave a Comment
37 Responses
  1. Avatar Michael says:

    The observation on the growth of cross-talk issues in FTTC I feel will become more important in the world of BDUK funded projects in the coming years. If councils only measure or have agreed contractually to measure minimum line rate at lets say 30Mbps down when cabinets are commisioned, but have not ensured that at end of contract period the rates must not have declined due to increased take-up/cross talk in the bundle then I can see lots of upset people. Or perhaps to mitigate this up to a point Openreach will have no choice but to deploy vectoring to maintain minimum guaranteed headline line speeds.

    1. Avatar GNewton says:

      Let’s go to the pub and buy a pint of ‘best efforts’ beer, if you are lucky you get half a pint 🙂

  2. Avatar No Clue says:

    What terrible latency on BT FTTC, must be suffering worse congestion than what the BT fanatics tell us Virgin suffer from.

    1. Avatar Ignitionnet says:

      Some of it will be accounted for by interleave, other parts the ‘interesting’ routing across the BT Wholesale network.

      Congestion will likely be excluded as an outlying fault.

      VM do have some localised congestion issues on their network, no question.

    2. Avatar Ignitionnet says:

      Actually looking at the graph there’s no indication of congestion there at all. The 24 hour latency is identical to the peak time latency.

      Combination of BT Wholesale routing and perhaps some traffic management.

      Lemme test.


      Pinging http://www.linx.net[] with 64 bytes of data:
      1.Reply from Size: 64bytes Time:16ms TTL:54
      2.Reply from Size: 64bytes Time:16ms TTL:54
      3.Reply from Size: 64bytes Time:16ms TTL:54
      4.Reply from Size: 64bytes Time:15ms TTL:54

      Ping statistics for []:
      Packets: Sent:4, Received:4, Lost:0 (0% loss)
      Approximate round trip times in milli-seconds:
      Minimum: 15ms, Maximum: 16ms, Average: 15ms


      Pinging []with 64 bytes of data:
      1.Reply from Size: 64bytes Time:12ms TTL:60
      2.Reply from Size: 64bytes Time:11ms TTL:60
      3.Reply from Size: 64bytes Time:11ms TTL:60
      4.Reply from Size: 64bytes Time:11ms TTL:60

      Ping statistics for []:
      Packets: Sent:4, Received:4, Lost:0 (0% loss)
      Approximate round trip times in milli-seconds:
      Minimum: 11ms, Maximum: 12ms, Average: 11ms

      Plusnet route is considerably more direct.

    3. Avatar No Clue says:

      I wonder why my BT connection is…

      Pinging linx.net [] with 64 bytes of data:

      Reply from bytes=64 time=9ms TTL=53
      Reply from bytes=64 time=8ms TTL=53
      Reply from bytes=64 time=8ms TTL=53
      Reply from bytes=64 time=9ms TTL=53

      Ping statistics for
      Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
      Approximate round trip times in milli-seconds:
      Minimum = 8ms, Maximum = 9ms, Average = 8ms

      Random congestion on BT…

      Many would be way better of with lower peak latency and higher speeds on Virgin.

    4. Avatar FibreFred says:

      Get on the forum and tell them to switch the No Clue if the grass is greener

      Good to see Ignition putting your poor assumptions straight


    5. Avatar Steve Jones says:

      Pinging linx.net on my BT Infinity connection I get 9ms (on all 4 pings).

      So no sign of congestions on this Saturday lunchtime. I’ve actually not seen much variation

      The minimum ping time is really a function of the routing (and interleaving). What I see from my location (in Maidenhead) is no real guide to what you might get from another location, so it’s difficult to generalise.

      I’d be most concerned about how representative the Ofcom stats are, given that it’s a small sample size which is largely self-selecting. They may have done some re-weighting of the numbers to try and make it more nationally representative, but I have grave doubts as to it’s accuracy.

      Of course there is a much better way of obtaining what the (potential) speed of ADLS and VDSL lines are, and that is for the SPs to collect the sync speed stats as all that information is collected at the DSLAM (and, in BT’s case at least, is used to automate management). There will also be stats about drop-outs, re-syncs, attentuation, noise margin, potential sync rates and much else.

      If the SP’s copuld be required to provide such stats, then we’d at least have a more accurate country-wide image of potential for the lines. Contention is a different matter (although the SPs should be able to characterise that too).

      I realise this stuff would be commercially sensitive, but even (anonymised) national or regional pictures would be surely be helpful.

    6. Avatar No Clue says:

      Ignitionnet has actually confirmed things his ping time is almost double that of mine and steve jones above. His plusnet connection pings a third faster also. That combined with BT forum posts and the Ofcom stats all point to BT being congested. Terrible product good to see Virgin who you usually bash waaaay ahead of them now.

    7. Avatar Steve Jones says:

      There is no evidence of congestion. The characteristic of that is a lot of variation in ping times, especially between known busy and quiet times. It’s generally all about routing (and geography). There there are the different peering arrangements of different SPs of course. Depending where you are in the country and who your operator is, that can vary a lot. It can also vary according to dynamic changes in IP routing. Another factor is redirection or different address resolutions. These will often change according to locality. That’s especially true for streaming services.

      It’s always important to be careful about any measurements and what they purport to show.

    8. Avatar Ignitionnet says:

      There is no congestion on either of my services, they run at full speed 24×7. The latency is higher on the BT service because I’m further from LINX than you are and the route my BT line takes is indirect.

      The route could certainly do with being more direct but there’s no congestion evident on it. There was some congestion earlier in the year on the BT Wholesale network which was fixed. It affected both of my lines.

      The congestion on Infinity is due to BT Wholesale network issues, which affect multiple ISPs sharing bandwidth out of handover points with them. BT Wholesale underestimated the speed at which bandwidth usage would grow and were caught with pants down. They’ve had to install new hardware in a number of places along with software upgrades.

      To be fair to them they are not unique in this regard, Sky were also caught by surprise by bandwidth requirements – this exchange was also on their lists as congested and ran poorly for a similar length of time to BT Wholesale.

      As I said, there is no evidence for widespread congestion, if there were it would show as the latency on the 8-10pm graphic being considerably higher than the 24 hour one.

      Tracertingwww.linx.net[] ,Maximum hops:30
      1 5ms 13ms 5ms
      2 5ms 5ms 6ms
      3 19ms 8ms 8ms
      4 8ms 7ms 8ms
      5 8ms 8ms 8ms
      6 7ms 7ms 28ms
      7 17ms 16ms 16ms
      8 15ms 15ms 14ms
      9 17ms 17ms 17ms
      10 19ms 19ms 19ms
      11 19ms 19ms 19ms
      12 15ms 16ms 15ms
      13 16ms 16ms 16ms
      14 15ms 16ms 16ms
      15 15ms 15ms 16ms

    9. Avatar No Clue says:

      Nope your latency is double mine, and the same amount of hops to linx, not to do with time of day either both your tests and mine were within half an hour of each other…

      1 <1 ms <1 ms <1 ms 1xx.x.x.x
      2 7 ms 6 ms 6 ms
      3 6 ms 6 ms 6 ms
      4 7 ms 7 ms 7 ms
      5 7 ms 7 ms 7 ms
      6 8 ms 7 ms 7 ms
      7 8 ms 7 ms 7 ms
      8 7 ms 7 ms 7 ms
      9 8 ms 8 ms 7 ms
      10 8 ms 7 ms 8 ms
      11 8 ms 8 ms 8 ms
      12 8 ms 8 ms 8 ms
      13 8 ms 7 ms 7 ms
      14 8 ms 8 ms 8 ms
      15 8 ms 8 ms 8 ms
      16 9 ms 8 ms 8 ms

      Trace complete.

      Your latency is DOUBLE.

      This also backs up the ofcom data with the baseline/lowest Plusnet in the chart above figure being just below that of BT, just as it is recorded with your connection. (11ms vs 15ms).

      Also makes sense with regards to all the posts on https://community.bt.com/t5/forums/searchpage/tab/message?filter=labels%2Clocation&location=forum-board%3ABTInfinity&q=congestion

      I guess i am one of the lucky ones, not to suffer rubbish BT ping times.

    10. Avatar FibreFred says:

      Hahaha you really do need to stop now as you are totally living up to your current id ‘no clue’

      Not that I mind you being shown up , but …. do yourself a favour

    11. Avatar No Clue says:

      So you disagree his latency is double mine?

    12. Avatar Ignitionnet says:

      Can’t say I’ve ever been told my service has congestion by someone else before. Would’ve thought I’d have been in the best place to decide but evidently not.

      Congestion gets excluded from these stats as an outlier I believe. Heavy congestion is considered an atypical fault.

      The longer ping time on my BT service is accounted for by routing on BT Wholesale’s network, not congestion, simply because it doesn’t change throughout the day – you’d expect as load increases it to rise, it doesn’t.

      Given their market share it’s not surprising BT have more complaints of congestion than anyone else. Plusnet are #2 and their share is about 5%.

      In any case believe what you want to, I can only report my experience, and state what I’ve read from the experience of other ISPs, that BT have the same issues with congestion as the rest of the BT Wholesale cohort, and have somewhat higher ping times from the regions due to the nature of the routing on the BT Wholesale network.

    13. Avatar No Clue says:

      There are several addresses in the chain before and after traffic hits the BT core servers where your ping is double. Where it did hit BT servers (mid-ish way through) we appear to be routed very similarly. I therefore doubt “The longer ping time on my BT service is accounted for by routing on BT Wholesale’s network”.

    14. Avatar FibreFred says:

      Hey…. you know best about his connection, in fact you know more about networking in general than he does (let’s forget about GPON etc for now)

      So… yeah I guess Ignitionet is wrong this time 😉

    15. Avatar No Clue says:

      I never said he was wrong on anything i simply pointed out his latency is double that of mine, even when taking close to the exact same route.

  3. Avatar gerarda says:

    The sample now is now only 975 not 2175 making the report even more statistically suspect then ever

    1. Mark Jackson Mark Jackson says:

      No it all stems from the 2,175, although some specific tests are based on a smaller part of that sample. For example, the summary of average download speed by ISP package uses 1,711 as some data from connections that aren’t popular enough to be categorised, or not relevant to the specific test, doesn’t get included etc.

    2. Avatar gerarda says:

      The average speed data – which is the result you headlined – is from a sample of 956

  4. Avatar GNewton says:

    Is it just me, or has anyone else noticed that there is hardly any mention of the upload speeds. With no more than 1mbps on ADSL2+, this makes the latter virtually useless for many applications, such as cloud services, especially for business users. Even VDSL, with slighly better upload performance than Virgin cable, is not up to the task for many users in this respect.

    Fibre optic broadband should be available in symmetric speed packages. (VDSL is NOT fibre optic broadband!)

    1. Avatar fastman2 says:

      Gnewton Fibre optic broadband should be available in symmetric speed packages – oh that woud, be a leased line then !!!!!!

    2. Avatar No Clue says:

      Since when did a lease line have to be fibre? Another load of tosh.

    3. Avatar Ignitionnet says:

      I am sure someone would be happy to sell you a 10Mb symmetrical service delivered over FTTC.

      To describe the upstream as not being up to the task for many users seems an odd remark. I am not aware of many people who are using applications that require >19Mb upstream. The current speeds certainly make online backup take longer, however it’s enough to stream a 4K video outbound.

    4. Avatar No Clue says:

      If upload speed is important to anyone they would ironically be better off with a mobile connection. DTMark who posts on here uses a mobile connection and has regularly posted with upload speeds around 20Mb, ironically faster on some occasions than what so called “next gen fibre” from BT goes at. NO need for a leased line even if some tools think all them are fibre.

  5. Avatar fastman2 says:

    Leased lines are ymmetric speed packages – broadband is not

    1. Avatar No Clue says:

      What is SDSL then if it is not broadband?

    2. Avatar Gadget says:

      @No Clue – not widely available

    3. Avatar gerarda says:

      All broadband can be symmetric it is just more efficient (i.e. cheaper) to offer higher download speeds and restrict uploads. Over time it seems the services offered have got more asymmetrical but user complaints have been negated by a general increase in download speeds to compensate.

    4. Avatar No Clue says:

      “@No Clue – not widely available”

      Neither are leased lines, so your point is what? SDSL is available at these…
      https://www.samknows.com/broadband/exchanges/bt/sdsl which is around (cant be bothered with you to be exact) 20%. SDSL likewise is copper based and not fibre based so that debunks your other idiot ID nonsense that its only leased lines that are symmetrical and its only fibre that is symmetrical.

    5. Avatar Gadget says:

      @No Clue – you seem to be mis-informed about the availability of leased lines, especially compared with SDSL. Openreach availability for leased lines is found here http://www.openreach.co.uk/orpg/home/products/ethernetservices/whyor/ournetwork/ournetwork.do and states “We have an access presence in over 5,500 telephone exchanges nationwide “, compared with SDSL the “around 20%” [your estimate]for SDSL.

    6. Avatar Abuse Alert says:

      Any comments with personal abuse and the words like ‘idiot’ should be removed. Why are some incapable of a polite discussion?

    7. Avatar No Clue says:

      And SDSL is available to 898 exchanges so the guesstimate or as i stated “cant be bothered with you to be exact” of 20% was not that far off considering 20% of 5500 (THATS THE TOTAL OF EXCHANGES NOT THE AMOUNT LEASED LINES ARE IN) is 1100. Oh shock i was 202 out with a guess, ring the authorities. Its as near as 16.33%, happier now?
      I like the way though BT not being accurate with the “over” 5500 claim but me quoting anything not 100% spot on is not OK LOL

      Run along.

    8. Avatar No Clue says:

      Oh and as for MR Abuse Alert ID who wants “polite discussion”…..

      Follow your own advice in future. Rather than as you would say “sadder with every one of your comments”

    9. Avatar Gadget says:

      Regardless of the way the numbers are expressed you appear to be missing the point that the previously posted comment of yours about leased lines “Neither are leased lines, so your point is what?” has been shown to be incorrect.

  6. Avatar TheFacts says:

    The number of exchanges with SDSL is not important, it’s the number of customers it’s available for that matters.

Comments are closed.

Comments RSS Feed

Javascript must be enabled to post (most browsers do this automatically)

Privacy Notice: Please note that news comments are anonymous, which means that we do NOT require you to enter any real personal details to post a message. By clicking to submit a post you agree to storing your comment content, display name, IP, email and / or website details in our database, for as long as the post remains live.

Only the submitted name and comment will be displayed in public, while the rest will be kept private (we will never share this outside of ISPreview, regardless of whether the data is real or fake). This comment system uses submitted IP, email and website address data to spot abuse and spammers. All data is transferred via an encrypted (https secure) session.

NOTE 1: Sometimes your comment might not appear immediately due to site cache (this is cleared every few hours) or it may be caught by automated moderation / anti-spam.

NOTE 2: Comments that break our rules, spam, troll or post via known fake IP/proxy servers may be blocked or removed.
Cheapest Superfast ISPs
  • Hyperoptic £22.00
    Avg. Speed 50Mbps, Unlimited
    Gift: None
  • Onestream £22.49 (*29.99)
    Avg. Speed 45Mbps, Unlimited
    Gift: None
  • xln telecom £22.74 (*47.94)
    Avg. Speed 66Mbps, Unlimited
    Gift: None
  • Plusnet £22.99 (*35.98)
    Avg. Speed 36Mbps, Unlimited
    Gift: £50 Reward Card
  • Vodafone £23.00
    Avg. Speed 35Mbps, Unlimited
    Gift: None
Prices inc. Line Rental | View All
The Top 20 Category Tags
  1. BT (2769)
  2. FTTP (2756)
  3. FTTC (1785)
  4. Building Digital UK (1743)
  5. Politics (1667)
  6. Openreach (1622)
  7. Business (1434)
  8. FTTH (1340)
  9. Statistics (1241)
  10. Mobile Broadband (1227)
  11. Fibre Optic (1064)
  12. 4G (1057)
  13. Wireless Internet (1021)
  14. Ofcom Regulation (1015)
  15. Virgin Media (1006)
  16. EE (697)
  17. Vodafone (669)
  18. Sky Broadband (668)
  19. TalkTalk (663)
  20. 5G (517)
Helpful ISP Guides and Tips

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved - Terms , Privacy and Cookie Policy , Links , Website Rules , Contact