Broadband News

UK could be number 1 in broadband league tables rather than 35th

Broadband in the UK has since the days of Pipex only being popular when it was cheap, with millions valuing saving a couple of pound a month versus the highest speeds, combine this with the late start of the VDSL2 roll-outs and alternate providers only starting to deliver full fibre seriously in the last year or two and it is no surprise that in some comparisons the UK does not fare as well as other countries.

Cable.co.uk has run its regular analysis of the open data available from the Mlabs speed test and declared that the mean download speed in the UK is a lowly 18.57 Mbps and at position 35 and almost the last of the EU countries. Many headlines have featured the ranking of Madagascar at position 22 with a 24.87 Mbps mean. 

The 18.57 Mbps is massively lower than the Ofcom average speed for the UK which was 46.2 Mbps and our average for Q2 2018 is 30.7 Mbps We don't know exactly why the Mlabs figure is so low, it is possible that they are seeing an even higher number of Wi-Fi based devices than we do, we know that if we exclude mobiles and tablets the UK mean jumps to 34.8 Mbps in Q2 2018.

A great many of the comments presented by full fibre operators in response to what is generally being represented as a failure of the UK broadband industry is that we need to roll-out millions more homes of full fibre and very quickly, and while we welcome this sentiment there is a much lower cost route and one that could be delivered in a number of weeks.

So how can the UK rapidly climb the broadband league table?

The answer is simple and it simply needs Virgin Media to do something like sell a minimum 100 Mbps service, i.e. move all those customers on the 50 Mbps and 70 Mbps service onto their 100 Mbps tier, the effect of just upgrading existing Virgin Media cable customers would add around 2 Mbps to the mean figure for the UK, shifting us up to around position 28 in the league table. A more radical upgrade to make a 200 Mbps the minimum speed sold by the cable operator would add around 15 Mbps to the mean, and propel the UK into the top 10 just ahead of Jersey. In terms of numbers of lines this would mean around 4 million broadband connections would be on a 200 MBps or better service out of the 25.3 million fixed line broadband connections.

So there is the proof in a way, that due to the public voting with their wallets and not buying even a mid tier speed when speeds when 300 Mbps and faster is available that we are not moving up the speed tables.

This reality is actually a cautionary note for those pushing full fibre as the answer to all broadband woes, i.e. if you sell multiple speed tiers making Gigabit available to millions may not have a big impact, to have an impact the millions of full fibre lines promised in the next couple of years need to have minimum speeds of 100 Mbps or 200 Mbps.

This pattern is likely the explanation for why some countries are doing so well, i.e. their minimum cable package and full fibre speeds sold may be higher than the UK, or the choice is between just ADSL2+ and full fibre, rather than what we are going to see in the UK which is going to be a choice of ADSL2+, VDSL2, cable and full fibre and the slower services are going to continue to be popular due to price points of £15 to £20 per month.

While running up the spreadsheets to see what effect a shift of Virgin Media customers would have, we run some other figures through the spreadsheet and if the product split was combined with the average speeds that are used in broadband advertising we got a mean download speed of 47.8 Mbps, with an initial shift to a 100 Mbps minimum for cable services this rises to 51.3 Mbps and for a minimum 200 Mbps increases to 65.4 Mbps.

The fact that the 47.8 Mbps is so close to the 46.2 Mbps from Ofcom, is largely a factor that the Ofcom data is based on the same Samknows system as many of the large providers are using for their advertised speeds, but it does show everyone that the spread of products we see from the public is very similar to the model used by Ofcom to arrive at its UK average figure.

Another scenario we have explored is a full conversion of people on ADSL/ADSL2+ to the entry level VDSL2 service (40/10) for the various providers and no changes to product speeds for Virgin Media and the effect on our UK speed test average would be a rise from 30.7 Mbps to 34.5 Mbps and the Ofcom equivalent figure would be around 54.7 Mbps. Using the 3.8 Mbps improvement in our results, if the Mlabs results showed the same it would move the UK up to around position 28.

Combining the ADSL2+ to VDSL2 and a 200 Mbps minimum for cable services would lift our average to around 53.3 Mbps and the Ofcom equivalent to around 75.4 Mbps and potentially lift the Mlabs figure to something like 41.17 Mbps lifting the UK to a potential top 3 position, add in some of the FTTP and G.fast roll-outs already underway and a number 1 position is possible.

For those saying full fibre is the way to climb the league table, consider the plight of Korea and Japan with averages of 20.63 Mbps and 28.94 Mbps respectively which are often used as examples of massive full fibre availability and you can see that the mean is surprisingly stubborn, hence us exploring a different path if there is a political emperative and economic to climb up the global broadband speed tables.

Conclusion Yes more full fibre is a good move, but there are lots of other ways to improve things in the UK and we need to take a long look at how long ADSL/ADSL2+ MPF services are supported. 

Warning If we focus purely on climbing the league tables, then services such as G.fast which can be rolled out to millions with minimal investment compared to full fibre are going to take precedence over full fibre, i.e. the decisions made in the 2008 to 2012 period are ripe for repetition.

If you want to see the full list of countries ISPreview has reproduced the full list of 200 countries, we have skipped the table as we wanted to focus more an analysis of what can be done to change the UK result.

Comments

The fastest and cheapest way to increase the UK's average speed would be simply to switch off all ADSL/ADSL2+ connections. A lot of people would lose their broadband connection, but the average speed would increase, and we'd move up the table. We probably shouldn't fixate on such league tables.

  • sheephouse
  • 10 days ago

Provide it, and we'll buy it. I pay £42/month for 8Mb. Yes, 8Mb. I would happily pay £100 or more per month, if it got me good speeds (80Mb+). But I can't buy it because my exchange doesn't provide it. And even £100/month wouldn't necessarily be my cap.

"Built it and they will come" is often quoted. Provide the service and people will buy it. But if there's no service there, we can't move up the rankings.

Price isn't everything, when 8Mb is your only choice......

  • jimwillsher
  • 10 days ago

Like the UK Government this is just a numbers exercise. No one is interested in getting decent broadband to those who don't have it. Just display some crap statistics to get up a fictional ladder.

  • galacticz00
  • 10 days ago

@jimwilsher and if they offered you 40/10 for £30 per month would you buy that? Would you then pay £100 a month for 100/20? A small number of people might buy the higher figure but the majority would just go for the 40/10 because it is good enough and lower cost. That is the reality in the UK - a small percentage will pay more for a significantly higher speed, most will pay a moderate amount for a moderate speed (unless that moderate option is not offered at all).

  • ian72
  • 10 days ago

"Price isn't everything, when 8Mb is your only choice......"

8Mb would be a 4x improvement for me! Stuck on a 20CN exchange so bonding lines is not an option thanks to no ISPs that support it taking new orders on 20CN lines. Although a VDSL cab is being installed, due to distance from the cab, an upgrade is not available or viable.
Full fibre IS the way forward and league tables be damned.

  • nrb501
  • 10 days ago

For those saying they are stuck on ADSL with no better option, reality if the press coverage of the league table is allowed to influence things would be that the money that could boost you to say 30 Mbps could be used to boost a greater number of people in urban areas from 30 Mbps to 200 Mbps or more and thus have a bigger effect on the mean and position in the league table.

Or put another way, the past policies of delivering something better to as many people as possible and is already shifting towards a speed boost mentality is largely why the UK is where it is.

  • andrew
  • thinkbroadband staff
  • 10 days ago

The large effect of shifting the Virgin customers simply confirms that it is a nonsense to use mean average instead of median when you have a mix of technologies at hugely differing speeds.

  • gerarda
  • 10 days ago

Also I'm stuck with 3meg ADSLmax, with no other option, cab was upgraded in 2014 but it's 4km away, the recent 3km infill cab no use either. The superfast figure for the area being 80.2%.

Boosting the speed availability of those who already had the best speeds has meant people haven't upgraded as they where already happy with the better speed they already had.

Looks like a straight average is not an adequate way to measure peoples access to the internet. Even median speed would be better.

  • brianhe
  • 10 days ago

Hence why we usually include both mean and median on our sets of statistics

Though an interesting point is that the median could be 200 Mbps and higher if ultrafast speeds were ultimately important to the millions passed by Virgin Media and other ultrafast services.

Median speeds have an odd stepped behaviour sometimes, you can see this with the upload speeds for some providers in our various monthly results sets

  • andrew
  • thinkbroadband staff
  • 9 days ago

Even if you shifted everyone to a faster product, if they are still speed testing via WiFi or a mobile device then it may not make as much difference as you think.

I have Vivid 350, but I don't get very good speed test results on mobile devices, on desktop pc I get 381Mbps down.

This is perhaps why Ofcom figures are higher, Samknows boxes are all hardwired I believe, and perhaps more tech savy people have them, and thus faster connections.

  • R0NSKI
  • 9 days ago

I agree with the comments made here, take the US for instance where I am right now... For me to get a decent TV package, at a mid-tier level, the slowest broadband I get with any of the bundles is 150Mbps which is quoted as "good for a small family" and "up to 8 devices." The UK has a history of relatively low prices compared with e.g. the US and Australia, this causes users to focus on a £10 difference, whereas in America for instance the price is $100, hence when you see $110 for twice the speed you jump for it. When users are paying £25 in the UK, a step up to £35 seems steep for 2x speed.

  • ukhardy071
  • 9 days ago

The marketing by companies is very significant, where XFINITY in the US market 100 and 150Mbps as a mid-tier grade, a standard home with 1 or 2 kids think that is the minimum they can opt for.

Compare that with the UK where superfast is advertised on a 40Mbps product, users believe that will cover their family of 6 just fine.

To give an example, virginmedia have "superfast" and "ultrafast" on the 50Mbps package online.

On Xfinity in the US they have, "25mbps to 60mbps. Good for 5 devices, light streaming, downloading music and photos."

As a consumer, you get a very different idea.

  • ukhardy071
  • 9 days ago

@R0NSKI Already covered and hence why we have different figures for our speed test projection and an Ofcom one, i.e. our projection is based on average seen already for the Virgin 200 Mbps service, so will include Wi-Fi effects, the Ofcom one is based around the advertised speeds which are meant to be a median peak time speed for an Ethernet connected device.

  • andrew
  • thinkbroadband staff
  • 9 days ago

Virgin had dropped the 50 Mbps tier but re-introduced it at a price point to compete with the VDSL2 services it seems, i.e. while speed is key for some, it appears that for many the price outweighs the speed factor

  • andrew
  • thinkbroadband staff
  • 9 days ago

I agree with Andrew's conclusion... People are buying the cheaper product rather than the fastest available. That naturally curtails demand for even faster technology to be deployed.

Just look at the comments on here... Predominantly from people on sub-superfast demanding modest improvements.

Matt Hancock did the country a disservice by stopping the BDUK focus on the final few and switching to undermining BT on full fibre.

Nuff said.

  • WWWombat
  • 9 days ago

@ukhardy071, you speak as if the services are uniform across the US. If it is anything like it was a few years back then the provider market is fragmented and the offerings in one area can be very different in speed and price to another. The UK is more of a uniform area with a very large percentage having access to similar products (even if you just took Virgin then over 50% of premises in the UK can get their products).

  • ian72
  • 9 days ago

@Andrew I was referring to your surmising of bumping us up the league table by increasing broadband package speeds, if users are still testing by WiFi it wouldn't make much difference.

  • R0NSKI
  • 9 days ago

And that was accounted for in the stats, since for the league table changes our figure of 146 Mbps mean for VIVID 200 was used, rather than the advert figure of 213 Mbps

Our figure is from PC, mobile and tablet devices. Removing mobile and tablet does increase the speed some more, as does then removing the likely Wi-Fi tests for PC, but when removing PC likely Wi-Fi tests you also often end up excluding people in congested areas.

  • andrew
  • thinkbroadband staff
  • 8 days ago

Our exchange is still 20CN providing ADSL1 so what does that say when others are shooting the breeze about a 100Meg compared to a 300Meg service?

  • 21again
  • 8 days ago

@21again it says it is more economical to give people high speed upgrades when they already have the lines capable of taking it than it is to upgrade people who would require completely new installations at potentially high costs with little in the way of return. It also says the government are more interested in stats and league tables than individuals

  • ian72
  • 8 days ago

Reminds me of that Star Trek episode when spock says "Logic dictates that the needs of the many outweigh the needs of the few"

  • 21again
  • 8 days ago

I think it needs paraphrasing - "The profits for the shareholders outweigh the needs of the few, or the 1 (percent)"

  • ian72
  • 8 days ago

@ian72 yes absolutely. I would happily pay £100/month for 100/20. I work from home (IT technical consultant for an ERP), and I don't have other "common outlays" such as Sky TV, gym membership etc. Yes I would pay the £100.

If I were offered gigabit, I'd be prepared to pay £250/month. Yup!

Clearly I'm probably in the minority, since a lot of people want to get a great service for a fiver, but it's all about choice, and at the moment I don't have that choice.

  • jimwillsher
  • 8 days ago

@jimwillsher - I'm with you in being willing to pay for decent broadband, but having the opportunity (my BDUK rollout date has been moved back again).
There is a problem with things (in general, not just broadband) being too cheap. With a race to the bottom everyone is only concerned about price. When things cost more people start to think about quality more - value for money is about more than the lowest price.

  • sheephouse
  • 7 days ago

@sheephouse My Superfast date is showing as July to Dec 18. It previously showed as Jan to June 18, and previously it was "out of scope". Apparently my postcode is in the Gainshare category.

And I completely agree re: value. You get what you pay for.

  • jimwillsher
  • 5 days ago

@jimwilsher @sheephouse "Value - You get what you pay for"
If this was the case I would be paying pennies for my broadband of 0.5Mbs download speed. I'm aware that it is not pro rata and that there has to be a minimum charge for the connection and service provided, but there should be some level of proportionality.

  • sbeck201
  • 5 days ago

With the old ADSL/ADSL2+ services there was no differentiation.

On VDSL2 there is several speed and price points so there is a level of proportionality and one provider offers 15% back if the VDSL2 speeds do not sync at their guaranteed speed.

  • andrew
  • thinkbroadband staff
  • 5 days ago

> With the old ADSL/ADSL2+ services there was no differentiation.

I think Plusnet still have a lower price cost for the end user if they are in a "low cost area" and that is usualy a Market B 21CN exchange which can provide faster speeds using ADSL2 AND ADSL2+
I guess PN are the exception don't know of other ISP's who do this, but does it still cost the ISP more to provide ADSL1 from a 20CN exchange compared to ADSL1, ADSL2,ADSL2+ from a 21CN exchange, I think it used to?

  • 21again
  • 5 days ago

To clarify - no speed price differentiation on different speeds for the same technology i.e. 7.1 Mbps IPStream Max cost the same as 0.1 Mbps IPStream Max

75 Mbps VDSL2 costs more than 32 Mbps VDSL2 connection

  • andrew
  • thinkbroadband staff
  • 5 days ago

I agree with @jimwillsher and others about getting what you pay for, but not having the choice to do so. I'm with Talk Talk at 9Mb down & 0.9 up and would be happy to pay to get better upload speed because of my activities. However, Virgin have cabled my town so BTOpenreach is not bothering to compete and the exchange is not enabled for anything better. THis in turn denies Talk Talk and others dependent on BT O-R from competing with Virgin.

  • jonesjh99
  • 5 days ago

And in answer to the inevitable ripost: I don't particularly want to swith to one of Virgin's many, many package offerings or have a cable laid from the street across the top of my lawn to an entry point that is not convenient. BT have a perfectly good cable 3 feet down entering my house where I want it. I just need BT O-R to upgrade the speed to the cabinet, 30m away!

  • jonesjh99
  • 5 days ago

@jonesjh99 "THis in turn denies Talk Talk and others dependent on BT O-R from competing with Virgin" Not at all, there is nothing to stop TalkTalk investing some of their cash and installing their own network, something they have said they will be doing. As for BT not bothering to compete, I'm sure they would love to compete however I suspect BT's problem is that their bean counters have told them that it is currently uneconomic to do so.

  • MCM999
  • 4 days ago

Post a comment

Login Register