» ISP News, Key Developments » 
Sponsored Links

Openreach Put the Brakes on Future UK G.fast Broadband Plans

Wednesday, Sep 11th, 2019 (5:21 pm) - Score 38,088

Openreach (BT) has today informed ISPs that they are holding off giving any further guidance on future build plans for their hybrid fibre G.fastultrafast broadband” (100Mbps+) technology, which means that under the existing rollout they will only cover 2.73 million UK premises by March 2020..

The writing has been on the wall for awhile, not least since the last G.fast (ITU G.9700/9701) related rollout update was in November 2018 (here) and there have been no further updates (i.e. no new locations announced in 2019). The most recent coverage update confirmed that they had so far put the service within reach of 2,166,000 premises and the rollout pace was slowing (here).

What is G.fast? FTTP takes a high-capacity optical fibre cable all the way to your home, but in a G.fast setup the fibre only goes as far as your local PCP street cabinet and then runs to your home via an existing copper line. G.fast can make use of 106-212MHz of spectrum and thus delivers its best speeds for those within c.200-300 metres of a cabinet.

Under the original plan Openreach had intended to push the 330Mbps capable G.fast service to reach 10 million UK premises by the end of 2020 (March 2021 financial), although pressure from competitors, Ofcom and the Government’s move to prioritise “full fibre” networks has combined to encourage a greater focus toward much more desirable Fibre-to-the-Premises (FTTP) lines.


In August 2018 this resulted in Openreach’s decision to revise down their G.fast rollout plan to just 5.7 million premises by March 2021 (here), while today’s decision means that no new locations are currently planned to be added to the build programme. As such they will continue to rollout until March 2020, when just 2.73m premises will have been covered.

openreach q1 2019/20 network coverage and takeup

Just to be clear, the G.fast rollout hasn’t yet stopped but any build plans beyond March 2020 are currently under review. A final decision on that review is expected to be communicated to ISPs by March 2020 but we’d be very surprised if that included a major expansion beyond the planned 2.73m premises (probably not to the previously expected 5.7m level).

As before the reason for all this stems from Openreach’s desire to focus on FTTP deployments and avoid overbuild, which is a positive. Lest we forget that last year’s move to scale-back the G.fast rollout was ultimately followed by the operator’s decision to raise their full fibre rollout to cover 4 million premises by March 2021 (up from 3m), including an ambition for 15 million by around 2025 (up from 10m) – here.


Admittedly we suspect that the mixed performance and limited take-up of the G.fast service itself, which only tends to be very good over shorter copper lines, probably hasn’t done much to help its future prospects. By comparison FTTP can easily handle multi-Gigabit or better speeds in the future and doesn’t suffer from the same reliability issues, plus take-up has been good.

The BT Group are currently in the process of restructuring and a lot of that focus has to do with raising money in order to fund their full fibre plans. Today’s predictable news very much plays into that effort.

Tags: , ,
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 X (Twitter), Mastodon, Facebook and .
Search ISP News
Search ISP Listings
Search ISP Reviews
51 Responses
  1. Avatar photo Declan McGuinness says:

    Great news!

  2. Avatar photo CJ says:

    Does this mean they intend to roll out in more of the November 2018 locations after all, to reach the March 2020 guidance? It was starting to look like most of that list had.been abandoned.

  3. Avatar photo Deltoid says:

    Should have done that in the first place, rather than messing around pushing technologies that were niche and didn’t offer the coverage that FTTP would

  4. Avatar photo Bill says:

    I find this article quite misleading. BT have announced they are holding off giving further guidance.

    That is not the same as implying the plans have changed.

    Of course I would be quite happy for gfast to be shelved altogether and the resources directed to fttp.

    1. Mark-Jackson Mark Jackson says:

      You mean like in the very first paragraph, where this article says, “they are holding off giving any further guidance” and then later where it says, “Just to be clear, the G.fast rollout hasn’t yet stopped but any build plans beyond March 2020 are currently under review. A final decision on that review is expected to be communicated to ISPs by March 2020.

      I think that’s fairly clear and correct (it comes from Openreach). But from talking to ISPs and other sources I don’t know anybody who expects the existing plan to stay the course, at the very least their rollout will be revised down again or shelved next March. None of this is a surprise.

  5. Avatar photo Asrab Uddin says:

    I have a GFAST pod out side my house – but unable to order because of the route the cabling takes, physically its less the 80 meters but the cable length is 240 meters according to open reach engineer’s

    Being a housing estate out of the 50+ houses only 10 able to order what a waste of time, money and resources

    1. Avatar photo CJ says:

      Thinkbroadband have estimated both the number of properties that can get g.fast and the number of properties that are connected to a g.fast cabinet. On average, 60% of properties on the cabinets with live g.fast are within range. Your cabinet looks like it was selected by mistake if a long cable route means only 10 properties are within range.

    2. Avatar photo Kits says:

      Yes I live 23 houses from the cabinet and G.Fast stops two doors before me. Total waste of money.

    3. Avatar photo Mark says:

      It’s always a waste of money / a useless service if you can’t benefit from it. Not on the other hand, for the other 21 houses to the side of you.

    4. Avatar photo Optical says:

      Same problem here,cable doubling back on it’s self adds around a extra unnecessary 200mtr to my distance.

    5. Avatar photo Rubbishman says:

      Neighbors house is opposite mine has Gfast, but I can’t have it, but yet the phone line comes from the same post. Where is the fairness? Total utter rubbish, and we pay same price.

  6. Avatar photo adslmax says:

    That’s odd as they still doing G.fadst pods in the Cuckoo Oak area with more additions.

    1. Mark-Jackson Mark Jackson says:

      Try reading the full article max.

  7. Avatar photo FibreBubble says:

    If you have waiters on a VDSL cabinet it is quicker and cheaper to fit a gfast pod rather than another VDSL cabinet.

    1. Avatar photo Kits says:

      G.Fast is very limited on distance you need to be almost ontop of the cabinet for it to work.

    2. Avatar photo Meadmodj says:

      I think the suggestion here is that if some customers (low percentage) people can migrate to the G.fast product it frees up circuits on the FTTC cabinet for those further away.

  8. Avatar photo chris conder says:

    The superfarce shambles. As predicted years ago. This good blog from 2012 explains it all well. https://neil-fairbrother.squarespace.com/blog/2013/7/1/gfast-a-high-speed-cul-de-sac

  9. Avatar photo Matthew says:

    I think if they had gone with FTTdp G.Fast it would of made a bigger impact then it has done only CityFibre and others that have made them rethink like they have.

    1. Avatar photo A_Builder says:

      If you are putting in fibre to the DP on every pole you may as well just put the fibre into a connectorised block and offer FTTP as a connectable on demand product.

      That way it is cheaper to deploy as you don’t need the power to the DP faff as everything on the pole is passive.

    2. Avatar photo ENS says:

      Are there any rules prohibiting aerial fibre deployment? That would mean that FTTdp could be cost effectively extended to FTTP as there would similarly be no requirement for driveway trenching and re-instatement. And significantly more elegant a workaround than VMs green trunk flexing along the fenceline.

    3. Avatar photo Ivor says:

      “Are there any rules prohibiting aerial fibre deployment? That would mean that FTTdp could be cost effectively extended to FTTP as there would similarly be no requirement for driveway trenching and re-instatement. ”

      There’s plenty of Openreach FTTP done aerially. The whole “FTTP is expensive as it means digging up gardens” argument was always a total nonsense in those places where the copper is already overhead, and in newer builds it’s either already built in or can be easily retrofitted through modern ducts.

      In my village there’s a mix of both – I’m on FTTC sadly, but the rest of the village is mostly aerial FTTP, and some estates that were built about 15 years ago have been retrofitted with underground FTTP.

    4. Avatar photo A_Builder says:


      I’ve seen a fair bit of aerial deployment on my travels.

      Agreed the whole “we need to dig everything up” was just a part of the max-credible-FTTP-price to back up the OR-begging-bowl strategy.

      There is no good reason why copper aerial cant be replaced with aerial full fibre. Yes there are wind stresses etc and so the reliability figures are poorer for aerial but the cost differential between aerial and burying everything is soooooo huge that only a nutter would insist that everything is buried where there is existing aerial infrastructure.

      as I think a lot of people have said before on here if you could just get the fibre to a connectorised block on the top of the pole then the final drop can be done with pre-packed pre connectorised fibres by the linesmen when a customer orders fibre. As far as I am concerned at that satus fibre would genuinely be available to the user.

    5. Avatar photo John says:

      I should copy/paste this comment as it is tiring repeating it.
      Same garbage constantly repeated with G.Fast…

      Should have been FTTdp, should have rolled it out to DP’s as planned, etc.

      Learn what a DP is and where they are before criticising the current rollout.

      OpenReach G.Fast was NEVER intended to be FTTdp, but rather FTTrn (remote node).

      As pointed out in a previous comment if you are going to the trouble of running fibre right to the DP then in the majority of cases it would be easier/cheaper to just install FTTP.

      The cost of getting a power supply to a DP can be (and in many, if not most cases would be) many thousands.
      The logistics in powering a G.Fast DSLAM by reverse power (power fed by users modems over spare twisted pairs) was no small task either.

      G.Fast was originally to be deployed by OpenReach from a node (underground junction box or pole top) deeper in the network than the cabinet but nowhere near as deep in the network as a DP.

      G.Fast from a PCP is cheap but misses a large proportion of users who are too far away.

      G.Fast from a DP would be expensive, inefficient, and in most cases be closer than necessary for full speeds.
      It would require hundreds of thousands of very small DSLAM’s and would be the last method of deployment anyone with half a brain or knowledge of the network would use.

      G.Fast from a Node would have struck a good balance between the 2.

      G.Fast and FTTdp shouldn’t be in the same sentence unless you’re discussing MDU’s, large flats, apartments, high rises.
      Most of the UK isn’t like this.

    6. Avatar photo Meadmodj says:

      All too late now as mass rollout plans really needs to be FTTP and the OR Fibre First and Altnet investment confirms is now economic. This OR announcement is their rain check on current Cabinet G.Fast plans. But there is no reason why in exceptional circumstances G.fast can’t still be used on MDU, small industrial estates or a solar powered ONT (fibre to the first pole) for farms where the cost or timescales of aerial or buried fibre would be prohibitive. The objective should be effective speed, resilience and timescales.

    7. Avatar photo Ivor says:


      Openreach was very much considering the idea of FTTdp – indeed they showed off Netcomm’s equipment at Adastral Park.


      This variant only has the capability to handle a small number of lines, so it would need to be installed at every DP.

      This kit has been placed into full service in Australia’s NBN – including the reverse power element. Seems to have worked well for them.

    8. Avatar photo Meadmodj says:

      Most manufacturers now have a range of ONTs with G.fast out in 1 port, 8 port and 16 port including Nokia. Worldwide there are also Solar Powered UPS for these.

      Example of configuration options https://www.isemag.com/wp-content/uploads/2015/12/No_two_alike_2_1215_web.jpg (note date 2015!)

      I hope OR will be innovative on their approach. They can re-engineer to FTTP later.

    9. Avatar photo John says:

      Ivor that was a test, what’s your point?
      There isn’t much OpenReach won’t have tested at Adastral Park.

      Basically a media/publicity stunt to show the advantages and capabilities of G.Fast.
      The tech was initially planned for a node based rollout.

      The higher speeds mentioned in the article weren’t ever intended to be sold over G.Fast.

      Read the article YOU linked to…

      It describes OpenReach testing this kit AFTER changing from a FTTrn rollout to a FTTC based G.Fast rollout.

      The initial nationwide plan was from a node.
      Then later they decided on deploying from the cabinet.
      A series of FTTdp tests after going all in for an FTTC rollout doesn’t mean their plan all along was to install from the DP.

      When OpenReach announced the intention to use G.Fast they had no intention of going as far as every DP or to sell 1Gbps over G.Fast.

      For MDU’s or areas with direct buried cables or certain rural scenarios it might have made sense to install a few DSLAMs on DP’s.

      OpenReach have tested and installed a few mini node based VDSL2 DSLAMs.
      There are a number of mini Huawei VDSL2 DSLAMs across the country.
      That doesn’t suddenly mean OpenReach’s VDSL2 rollout plan was node based all along.

    10. Avatar photo A_Builder says:


      My recollection of being sent various bits of PR and being on BT investor calls is that it was floated that there were going to be pole mounted GFast pods. This was part of the stop-the-break-up-of-BT spiel because we can provide =stupidMb/s at the drop of a hat by magic tech approach.

      I agree the idea was pretty silly as the costs of the power provision would potentially be quite high – unless it could have been done via the ducts back to FTTC DLASM with a 50V combined with fibre as the PCP mounted GFast pods are now.

      BT Central may have intended Gfast from the node but they were saying telephone pole on the calls. Now that may have been because of the audience not understanding the difference between DP and node but that was what was said at the time.

      Personally I think killing the half arsed implementation of GFast that BT were rolling out is the best thing.

      I can see no downside in a good steady fibre rollout by OR.

    11. Avatar photo Ivor says:

      John, using that logic, everything except PCP based G.fast was merely a “media and publicity stunt”.

      And to be honest, it’s no surprise that FTTrn died out. The DP is a natural place to put something (and it can be reverse powered). So is the PCP – it already has power and fibre, though you lose the potential speed uplifts.

      Performing the necessary network surgery to insert randomly positioned “nodes” and run power/fibre to them was never going to be sensible, and would go against Openreach/BT’s stated position to move towards simpler networks.

      Openreach didn’t do it, but Australia did deploy what we would call “FTTdp” – and it doesn’t seem particularly “inefficient”!

      Your unnecessarily combative tone doesn’t really hide the fact that you’re no more of an expert than anyone else commenting here.

  10. Avatar photo AndrewH says:

    I don’t understand why you would roll out a tech that would need to be upgraded in the future rather than just go FTTP to start with as that won’t need upgrading for decades.

    1. Avatar photo Lee says:

      I agree vdsl2 should of been the last deployment, G.fast is only feasible if it is coming from a pole (as you have to be very close to it to get its benefits speed and not its problems from a long ish line)

      FTTP>G.Fast node that is on the pole is it self to save on cost/time of having to get access to the house to install the fibre to house and fibre to ethernet box inside (with pole based g.fast bt man only has to loop thorough the fibre G.fast module on the pole doesn’t need access to house) but adding G.Fast to a vdsl cab that only 5-20 people might use or can use seems a waste of time and money

      FTTP would of been nice from the start but it would of cost may times more and they still be installing it now and at the time there was no setup plans for consumer installs (1992 keeps poping back into mind if fatcher had not blocked bt from doing full copper replace to FTTP we be the in the top countries with full fibre network and gigabit Internet)

  11. Avatar photo Gaty says:

    So basically I’m stuck in the dark ages on 20Mbps until 2025 when hopefully BT will pull their finger out of their arse and actually provide a 21st century network. Wow. Bloody shocking. If I was OFCOM I would be fining BT millions every year they delay

    1. Avatar photo Ivor says:

      Fine them for what? BT isn’t required to deploy FTTP to everyone.

      The 21CN project was about modernising their core networks, and enabling better and more futureproof services, which of course it did.

    2. Avatar photo JamesP says:

      You are on 20mbps, GFast was never going to help you.

      FTTP is the way forward and it’s good to see they are now focusing attention on this technology. Hopefully they will work on bringing the sub-superfast areas up to speed while developing these ‘fibre first’ towns which generally already have Superfast.

    3. Avatar photo Brian says:

      You can console yourself you on still on a much faster connection than many, many thousands are still below USO.

    4. Avatar photo Bonjovi says:

      Glad to see a constructive and sensible conversation doesn’t stop cranks. Why would you fine a commercial organisation for following a strategy just because you disagree with it?

    5. Avatar photo Summer Is Here says:

      I don’t believe users who are on 20Mbps is considered a crime yet!!!!

    6. Avatar photo SuperFast Dream says:

      I wish I could get 20Mbps on a fixed line, try 3 – 4Mbps, now that’s a crime!

  12. Avatar photo CarlT says:

    These comments from Ed Richards, former Chief Executive of Ofcom, didn’t age well.


    Openreach could’ve done this so much better. It was dead as soon as they went for the cheapest, most expeditious PCP-based model.

  13. Avatar photo Meadmodj says:

    We are where we are. BT not only made bad network investment decisions but has also neglected the state of the copper network both maintenance and expedient growth, hence why the theoretical benefit of G.fast is simply not being met. For instance my home line does not go to the nearest cabinet but one that is further away. A situation which is not surmountable.

    The BT decision to switch OR to Fibre First finally arrived and now BT needs the freedom to target the surplus G.Fast supplies where best they can whether to cover a lack of FTTC capacity or business/MDU (within cabinet G.fast limit) where FTTP is impractical short term. I view this as a tactical position as they concentrate on planning their future FTTP rollouts.

    There will be some BDUK/USO FTTP provision but not much will occur until the Government confirm their intentions for rural. Those on rural ADSL or slow FTTC may have a long wait.

    For many of those on slow ADSL in urban should consider utilising 4g for the next few years while we get more visibility of who is coming their way.

    1. Avatar photo Nga for all says:

      If we include Cornwall, then the count of rural FTTP is 450k. In addition now Stratum NI will add c90k. the Welsh will add 23k. This quarter Cumbria, Kent, Lincolnshire, Leicestershire, Digital Scotland, Dorset, Suffolk and Essex are all adding FTTP at the rural edge.
      A few more announcements in the South West and there is a good chance of exceeding 98%.
      The money is within the BDUK process to complete this work.

    2. Avatar photo NGA for all says:

      …and if you look at the BDUK latest report there are a further 350k premises contracted yet to be delivered..

    3. Avatar photo Meadmodj says:

      All good if that is the case, especially FTTP but it still leaves 2% and although the maps may show Superfast coverage I suspect there are lines not performing as such due to various line issues. My point is simply that it is no good people complaining they will just have to be patient. But I would still like OR to be proactive if altering a route, repairing an existing cable or joint could resolve speed issues to a number of lines particularly if those lines are likely to be there for the next 6 years or more. They should have the network data to identify anomalies.

    4. Avatar photo NGA for all says:

      Meadmodj … cannot comment on the rate of change in Openreach, but if sufficient focus is retained on the BDUK rural funding, clawback and BT capital payments then it appears 99% is possible, date is resource dependent. This assumes fixes in suburban and urban are completed by the market, e.g G London has 120,000 premises that cannot get >30Mbps, which is 30k more than of all of Norther Ireland.

    5. Avatar photo Meadmodj says:

      Yes the Thinkbroadband map has RED dots nationally and even in G London some are down remote streets or in the leafy hills and downs. But with all the Altnet promises and activity London should be generally OK though and many of the 120,000 could avail themselves of now of an alternative e.g FTTC, 4G, VM, Public WIFI or piggy back on neighbours (BT WIFI etc) if only they were aware. Some could even afford FoD but don’t because their use is so moderate. Others will simply be down to the historic wiring, walk the streets of London and you will see old lead cables, stapled drop wire (yes) and any manner of terminals and jointed cables up the walls.

    6. Avatar photo A_Builder says:


      I think we are beyond the point of altering copper routes now.

      BT never really got into it.

      The amount of effort involved in altering a route would be getting close to FTTP levels of spend. With the present situation all spend should be devoted to getting the FTTP roll going.

      It is heartening to see rural FTTP still rolling at a perceptible rate using the recycled BDUK funds. As this tends to be the outliers who cannot get FTTC then this is making good inroads into the last few percent and the long tail problem…

      Now for an EO plan!!

  14. Avatar photo David W Smith says:

    Should have spent the time and money investigating exchange only lines.

  15. Avatar photo Z says:

    I think this is a pretty bad idea considering that rolling out G.Fast is a much faster and cheaper process than FTTP, meaning that it will probably take longer for most places to get ultra fast broadband now.

  16. Avatar photo Techman says:

    BT and Openreach are an absolute joke. It took them 4 weeks, 8 visits, 9 engineers and digging and re digging up the road 3 times to fix a line fault at my house and it turns out they didnt even need to dig as it was the connection to the pole that was faulty. They are absolutely incompetent.

    GFast was and still is completely pointless. Honestly hope BT go bust and get bought out by a company with brains.

  17. Avatar photo Keith says:

    we are lucky to get 1to2 mbps when it is working, we have no mobile signal either, yes we do live in a remote area but the next village has a fast connection and the other end of the valley 2 miles away has a fast connection, we are going to be in the 2 percent that end up with no better service, as far as the 4G goes we are not in an area that is covered by that, I can see it costing us a fortune to get any thing better.

  18. Avatar photo Mark says:

    The disparity for those on superfast and ultrafast at speeds exceeding say about 70Mbps is relatively small.

    What do you think will happen when FTTP arrives in the short term but those on those speeds will be served first in the queue because of the distance to the nearest fibre box being a short run of fibre.

    The cost of running fibre 2km and digging up roads all over the place? I cannot see FTTP being deployed by 2025 to every premise in the country that is just pie in the sky.

Comments are closed

Cheap BIG ISPs for 100Mbps+
Community Fibre UK ISP Logo
Gift: None
NOW £24.00
Gift: None
Virgin Media UK ISP Logo
Virgin Media £26.00
Gift: None
Vodafone UK ISP Logo
Vodafone £26.50 - 27.00
Gift: None
Zen Internet UK ISP Logo
Zen Internet £28.00 - 35.00
Gift: None
Large Availability | View All
New Forum Topics
Cheapest ISPs for 100Mbps+
Gigaclear UK ISP Logo
Gigaclear £17.00
Gift: None
BeFibre UK ISP Logo
BeFibre £19.00
Gift: None
Community Fibre UK ISP Logo
Gift: None
YouFibre UK ISP Logo
YouFibre £22.99
Gift: None
Hey! Broadband UK ISP Logo
Gift: None
Large Availability | View All
The Top 15 Category Tags
  1. FTTP (5724)
  2. BT (3573)
  3. Politics (2605)
  4. Openreach (2342)
  5. Business (2324)
  6. Building Digital UK (2278)
  7. FTTC (2061)
  8. Mobile Broadband (2044)
  9. Statistics (1830)
  10. 4G (1728)
  11. Virgin Media (1676)
  12. Ofcom Regulation (1498)
  13. Fibre Optic (1428)
  14. Wireless Internet (1418)
  15. FTTH (1383)

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