topic Re: FTTC Speed Issues - Which? Campaign in Plusnet Feedback

FTTC Speed Issues - Which? Campaign

danludlow — Thu, 18 Jun 2015 18:48:48 GMT

I've had a "Round Robin" email today from Which? Pointing out that only 1% of Subscribers to the PlusNet 76mbps fibre service are receiving those speeds. To advertise this package, they suggest, a minimum of 10% should be downloading at the maximum speed, and are asking for people unable to receive advertised speeds to sign a petition to force ISP's to advertise realistic speeds and then deliver the speed as "promised".
PlusNet are mentioned as being particularly poor in Which? opinion, the petition calls on advertising watchdogs to pull misleading speeds from advertising where they seem unjustified.
Whether people feel hard done by enough to complain is up to them. I am annoyed that my service speeds fell since taking up the service, but I'm not sure that guilt lies with PlusNet, so much As BTW/BTO?
If starting from scratch, you would not design the infrastructure in the manner of the existing phone system using long runs of copper wire would you, and would seek technology less prone to distance drop-offs and interference. Cabinets where I am, in the countryside, are much further away generally, than cabinets in town are from their consumers, but then in the sticks their are fewer consumers near the cabinets too, we are spread out.
However this issue runs, there is a lot of grumbling about broadband speed, and more will, I think, need to be done by BTO to get speeds generally where only very few currently (it seems) enjoy them.
Something to think about anyway.

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Thu, 18 Jun 2015 19:30:05 GMT

Already being discussed here http://community.plus.net/forum/index.php/topic,140611.0.html
But to answer the specific point about Plusnet - the total sample size was 2,000 for ADSL and FTTC together as stated here http://www.theregister.co.uk/2015/06/18/which_says_15_4_million_households_not_getting_promised_broadband_speeds/ so the actual Plusnet sample size is probably miniscule and totally unrepresentative
In addition it appears that they were testing using the Which speed tester and given the margin of error in speed test results the conclusion is dubious to say the least
Let us assume that aan FTTC customer is in sync at 80Mb/s giving an IP profile of 77.4 Mb/s and a Plusnet profile of 77 Mb/s
We know that the Plusnet profile system drops the speed slightly as well so they are looking at the difference between 76Mb/s and a bit less than 77Mb/s which is almost certainly well outside the accuracy of the speed tester
It should be noted that although the ASA guidelines refer to speed results and insist that 10% of the customers achieve that result there is actually no way that an ISP can verify that as they only monitor sync speed so I would assume that the up to 76Mb/s refers to sync speed
The other factor is that they just checked the speed test results for those Which subscribers without considering that the customers referred to may not be on the 80/20 product and even if they are the distance to the cabinet would preclude the achievement of the maximum speed

Re: FTTC Speed Issues - Which? Campaign

danludlow — Thu, 18 Jun 2015 23:47:56 GMT

Thanks for your considered and interesting reply OldJim. I am not surprised to hear that the sample was small and unrepresentative, I find that reassuring. I have been with PlusNet for some time now and am more than happy to remain so, it would have been all too easy to sign up to one of the many offer prices touted, and all too easy to lose out as a result, I feel.
Whilst my broadband is faster than many and adequate (so I probably shouldn't grumble), I find that the seemingly steady deterioration in speed over time, irks. I'm sure that its the technology in use with its flaws, especially seen over longer runs of copper, but I cannot believe that BTO did not know this from the outset and are looking at ways to overcome problems when common sense says that the solutions should already have been rolling out. To me, higher speed broadband will need fibre run much closer, to the pole, if not the premises, particularly in the countryside where cabinets are few and far between.

Re: FTTC Speed Issues - Which? Campaign

MattyC — Fri, 19 Jun 2015 09:11:41 GMT

This is probably as good a place as any to post our official statement in case any visitors read this post:

Quote
Plusnet takes adhering to industry standards and regulations extremely seriously. We meet the standards set by the ASA, Committee of Advertising Practice (CAP) and follow the Ofcom code of practice. When any customer is thinking of joining us we take them through the speed range for their individual line and clearly display this. Ofcom’s latest broadband performance report highlights our ADSL2+ speeds are also consistently faster than our competitors on a 24 hour average and at peak times when it matters most to our customers.

In my opinion, I think OFCOM are pretty thorough in making sure that ISP's quote estimated speeds to prospective customers. Sales agents have to read script informing of estimated speeds, and it's shown in our online provisions as well.
Matty

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Fri, 19 Jun 2015 09:30:07 GMT

To further clarify Matty's point
The speeds quoted are, I believe, connection speeds not speed test result speeds

Re: FTTC Speed Issues - Which? Campaign

danludlow — Fri, 19 Jun 2015 11:27:05 GMT

Thanks Matty and Oldjim for your clarifications. Please don't see this as a criticism, but simply making a point. When I signed up to Fibre Extra (80/20) last year, within days of it becoming available, I was given the estimated speed of 47Mbps (I haven't checked this at time of writing but the figure sticks in my mind). Actual speed was about 52Mbps for quite some time, but it deteriorated, as is perhaps inevitable? In February a speed above 35Mbps was felt impossible to achieve anymore and so my service was reduced to Fibre basic.
As I understand it, take up on my cabinet has not been spectacular, it was at roughly 25% seemingly, about 3 months ago, the cabinet came online about a year ago. I know that one of the ports at least is U/S because I was transferred from it after speed deterioration as it was generating errors, tens of thousands of them at times when it was more or less idling. This followed a severe local thunderstorm that knocked out the cabinet supply for about an hour one night. I would not be the least bit surprised to find that the card itself took a battering, and isn't at its best.
By switching back to my BTO Huawei modem and using my ASUS AC68U Router-modem as router only, I have seen the IP on BTW Performance Test to 40.7 from 37.4 (A profile it held for literally months) the underlying connection speed (at the PC, wired) has been 35.6Mbps Down, but yesterday that increased to 37.8Mbps after two Router resets (via software), and this morning that speed remains, cheering me up no end!
Where is all this going?
I feel that the 47Mbps expected speed when I signed up, is not only still possible, I think its still there, held back by the profile, OR limited by capping. Its just a feeling in my water. I have the impression that there are issues that could be resolved, but obviously not without effort and expense (I have had 5 engineer visits in the year I've been on Fibre for faults and speed, and that's not good business is it). I ask myself if there would be any point because my speed is more than adequate, and I can't really draw a conclusion, so probably it isn't.
One thing I dread is slowdown and or loss of service, if I knew current levels were reliable, I think I'd be able to be happy!

Re: FTTC Speed Issues - Which? Campaign

Kremmen — Fri, 19 Jun 2015 13:31:33 GMT

I don't fully understand all the hoo ha.
I've been with a few ISP's over the years and the advice is always to use the BT phone line checker to see what 'your' line is capable of. When I switched to FTTC for example the BT checker said 26/6 and 20/3 (impacted).
I'm impacted by alluminium and I'm getting 17/2 so I have no complaint as the engineer did say that one short stretch of ally was particularly bad but not being in ducting would not be changed till it failed completely.
I'm guessing people are using the 80/40 or 40/20 or whatever as gospel.

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Fri, 19 Jun 2015 14:23:07 GMT

this is definitely worth a read http://www.revk.uk/2015/06/broadband-speed.html

Re: FTTC Speed Issues - Which? Campaign

danludlow — Fri, 19 Jun 2015 15:31:57 GMT

@ Kremmen Hi Kremmen, thanks for your contribution. Just to clarify, I have never seen the "Upto" speeds as gospel, and I researched prior to FTTC availability. My predicted speeds have reduced since taking up the FTTC service, quite considerably. Also, the speeds that I now get on BTW Performance Tester would have shown red lines and "please contact your ISP to report a fault" until the start of this year. Now, almost any speed gets a green bar showing that its' within expectations.
The expectations are lowered to suit, I suspect that BTW/BTO were losing a lot sending engineers to faults, now they maybe live with them, or to be more accurate, their customers do.
For some time BTO tried to tempt me with offers (yes I know that PlusNet is part of BT), and when I looked they could offer Upto 80Mbps with expected speeds higher than I was achieving at the time. Cannot happen, can it. This was a sales ploy, pure and simple.
So no, I'm not misreading nor making mistakes, I was offered and sold a service almost exactly a year ago, that is now not achievable it seems. Isn't that what this is all about?

Re: FTTC Speed Issues - Which? Campaign

Kremmen — Fri, 19 Jun 2015 16:08:45 GMT

My comments wern't aimed at you, just an observation from many posts seen here.
What does annoy me is that the Gov think we are mostly on the 'broadband superfast highway' but the truth is far from that. As oldjim's link suggests a lot are too far from an exchange or fibre cabinet to get anywhere near the package speed. This 2mbps max upload for new customers is also far from the information superhighway and a knee jerk reaction thanks to OFCOM and the others meddling.
As more houses are connected it can only get worse.

Re: FTTC Speed Issues - Which? Campaign

g1000 — Fri, 19 Jun 2015 16:49:34 GMT

Quote from: Oldjim
But to answer the specific point about Plusnet - the total sample size was 2,000 for ADSL and FTTC together as stated here http://www.theregister.co.uk/2015/06/18/which_says_15_4_million_households_not_getting_promised_broadband_speeds/ so the actual Plusnet sample size is probably miniscule and totally unrepresentative

The data set is that collected for Ofcom's Samknows report. The numbers may seem small to the untrained eye, but for an ISP package to qualify for analysis in this report the number must pass rigorous statistical analysis. The recruitment is quite sophisticated and balanced with all this in mind in the first place too.

Quote
In addition it appears that they were testing using the Which speed tester

Nope, see above

Quote
The other factor is that they just checked the speed test results for those Which subscribers without considering that the customers referred to may not be on the 80/20 product

Nope, the numbers are package-specific

Quote
the distance to the cabinet would preclude the achievement of the maximum speed

There's the rub in my view, users get an estimate and this is surely what they are "promised", not the advertised Up to headline speed.

Quote
It should be noted that although the ASA guidelines refer to speed results and insist that 10% of the customers achieve that result there is actually no way that an ISP can verify that as they only monitor sync speed so I would assume that the up to 76Mb/s refers to sync speed

The ASA guidelines don't state that 10% of users actually have to record a speed test result. To defend a complaint, an ISP has to be able to support with evidence that 10% can achieve that speed result. My guess is records show at least 10% achieve sync of 80, and the ISP has, or can easily generate, a set of robust test results to prove that with that sync a customer can get a maximum download speed of 76 Mb/s.
Sorry for the geeky responses, I got bored and read the reports today.

Re: FTTC Speed Issues - Which? Campaign

g1000 — Fri, 19 Jun 2015 17:11:47 GMT

Quote from: MattyC
This is probably as good a place as any to post our official statement in case any visitors read this post:
Quote
Plusnet takes adhering to industry standards and regulations extremely seriously. We meet the standards set by the ASA, Committee of Advertising Practice (CAP) and follow the Ofcom code of practice. When any customer is thinking of joining us we take them through the speed range for their individual line and clearly display this. Ofcom’s latest broadband performance report highlights our ADSL2+ speeds are also consistently faster than our competitors on a 24 hour average and at peak times when it matters most to our customers.

Huh, I am going to challenge that as false and invite Plusnet to explain. The latest report is here: http://stakeholders.ofcom.org.uk/binaries/research/broadband-research/november2014/Fixed_bb_speeds_November_2014.pdf
The relevant results are on Page 18, fig 2.2:
"Significant differences, to a 95% level of confidence, between maximum, average and peak download speeds for ADSL2+ ISP packages: November 2014
No differences"

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Fri, 19 Jun 2015 17:20:06 GMT

Taking a couple of points from your post

You haven't disagreed with the 2000 figure
You state that the reference to 1% is package specific
So -
Removing those on ADSL and those on lower than 80Mbps products this must leave a very small number and that number cannot in any conditions be statistically significant when the fact that BT, TalkTalk, Sky and Plusnet (plus others) breaks the number down even more
As posted here - quoted - the ISP numbers are based on sync speed not data throughput and using Samknows numbers doesn't change that fact

Quote from: Oldjim
this is definitely worth a read http://www.revk.uk/2015/06/broadband-speed.html

and from the link

Quote
OFCOM stated that "BT’s ‘up to’ 76 Mbps package: Only 1% of customers received the maximum advertised speed.". For a start, that surprised me - and I suspect they are not actually looking at the line speed but perhaps measuring speeds of transfers in to the Internet (see below). I would expect that 10% of lines can get 76Mbit/s or more on that service. In fact I would say that 90% get what was advertised, which is "not more than 76 Mb/s" and 10% don't (i.e. they get more than 76Mb/s so the advert was misleading to them).

I should add that taken from the ASA document they always refer to speed not to the results of speed tests so it is fairly obvious (as Rev stated in the link) that the number refers to or is taken to refer to the connection speed

Re: FTTC Speed Issues - Which? Campaign

g1000 — Fri, 19 Jun 2015 19:08:14 GMT

Quote from: Oldjim
Removing those on ADSL and those on lower than 80Mbps products this must leave a very small number and that number cannot in any conditions be statistically significant when the fact that BT, TalkTalk, Sky and Plusnet (plus others) breaks the number down even more

Hi Jim, I have a lot of respect for you and your hard work for the Plusnet community. I don't want to argue with anybody, so I'll just put my points across, we might then have to agree to disagree
So, having checked, the sample size is in fact far in excess of that required to conclude that the proportion is <10% (p<0.01 let alone p<0.05). For Ofcom's comparisons, they all achieve statistical significance as stated in the report, all the methodology, intervals, graphs, the UCL professor who worked on them, all referenced in the report. If you wish to formally complain to Ofcom that these can't possibly be accurate, I'm sure they would investigate and report back.

Quote from: Oldjim
As posted here - quoted - the ISP numbers are based on sync speed not data throughput and using Samknows numbers doesn't change that fact
I should add that taken from the ASA document they always refer to speed not to the results of speed tests so it is fairly obvious (as Rev stated in the link) that the number refers to or is taken to refer to the connection speed

The guidance document is here: https://www.cap.org.uk/~/media/Files/CAP/Help%20notes%20new/speed%20claims.ashx
It is clear to me at least that for a claim of 76 Mb/s download speed, like Plusnet make, the speed should refer to download throughput at the application layer seen by the end-user, not sync speeds. Looking at ASA judgements, the ISPs indeed seem to submit speed tests as evidence not sync speeds, I saw one referencing Ofcom's report and one uSwitch speed test data.

Quote
Where advertisers make a numerical speed claim that is likely to be understood by consumers as the maximum speed of their service, they should be able to demonstrate that the speed is achievable for at least 10% of the relevant customer base.
Specific claims, such as “50Mb Download speed”, should be based on tests of protocols relevant to downloading large files.
Speed claims for a service in general should be based on speed testing that is representative of the activities that users generally perform.

My guess is ISPs think their numbers can comply, and they think they can prove it through calculation of the throughput after overheads, plus real testing on a sample of lines on max sync, which at least 10% of users have.
Maybe you can just ask Plusnet what 76 refers to, and how they arrive at it. If you are convinced they use sync speed not download speed, then in my view they are not complying with the ASA guidance. You can always ask the ASA to look into it, you will then get their view of what the number should refer to, and Plusnet's response on the matter.

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Fri, 19 Jun 2015 19:39:25 GMT

Very simply in the case of Plusnet the maximum theoretical data throughput is just shy of 77Mb/s and for any supplier on Btw supply just shy of 77.4Mb/s and under those circumstances I don't see any possibility that 10% of the user base manage to get 76Mb/s on speed tests.
So either we are comparing apples and oranges or somebody is fiddling
I just had a look at the Capacity Issues thread and I didn't see anybody getting a 76Mb/s x6 speed result
Going back to the sample size - where is the sample size for Plusnet stated
I can't ask Plusnet anything as I am just an ordinary user with no special access any more

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Fri, 19 Jun 2015 20:09:35 GMT

I just did a quick check
BT, Plusnet, TalkTalk, ID Net, Zen and Sky all say up to 76Mb/s and up to 38Mb's
From that I just don't believe that they are following the perceived advertising rules but are just making a theoretical calculation from the sync speed
The exception I found is Vivacity who say up to 80Mb/s http://www.vivaciti.net/fibre

Re: FTTC Speed Issues - Which? Campaign

ejs — Fri, 19 Jun 2015 21:06:49 GMT

Question about the ADSL2+ speeds statistics claims:
Does it help boost the stats if you provision the long lines (with estimates under 4Mb/s) as ADSL1 "up to" 8Mb/s by default? And then those slower speeds won't be included in the "up to" 24Mb/s figures, even though the figures are apparently adjusted to even out the effects of line length?

Re: FTTC Speed Issues - Which? Campaign

g1000 — Fri, 19 Jun 2015 21:27:26 GMT

Quote from: Oldjim
I just had a look at the Capacity Issues thread and I didn't see anybody getting a 76Mb/s x6 speed result
Going back to the sample size - where is the sample size for Plusnet stated
I can't ask Plusnet anything as I am just an ordinary user with no special access any more

But that is Which's very point! "Of the broadband packages we analysed, we found three that failed to meet CAP and BCAP’s guidance". That includes the only two 76 Mb/s packages analysed, BT and Plusnet.
How and whether Plusnet or another ISP can prove they meet the guidance for 76 Mb/s we simply do not know. I can only give my guess as to how they have all come up with that figure, as already described. I don't think it is impossible to get 76 Mb/s on the right speed test with 80 sync personally, so who knows.
As to 76 referring to sync, you can always try asking Plusnet as an ordinary user, surely a customer has a right to ask what 76 refers to, sync or download throughput. Maybe even ask the ASA, they may give a simple email reply confirming what it should be. I think I have done enough to prove it should be download throughput, and I have no reason to ask the question myself. Think of wifi, 4G, VDSL - the whole point is huge discrepancies in data rate overheads like that should be removed when advertising.
The sample size is stated in the Which report (n=209 for Plusnet 76 Mb/s), along with 95% CIs (upper limit still 1%), consistent with my own analysis stated above. I really mean no offence by this, but maybe you should try reading the report? I don't agree with their ideas/campaign of advertising one bit, but the facts and data are pretty thorough. I think maybe you rushed to judgement and made too many false presumptions on that one, sorry.

Re: FTTC Speed Issues - Which? Campaign

Oldjim — Fri, 19 Jun 2015 21:45:30 GMT

Since I don't subscribe to Which I can't
Why should I ask Plusnet - why don't you if you are so interested
In my post above I pointed out that virtually all the ISP's use the up to 76Mb/s which is a clear indication that they haven't done it via speed tests but by a calculation
If you read the link from the boss man at A&A you would see that he also seems to agree that the 10% refers to sync speed not speed test results

Re: FTTC Speed Issues - Which? Campaign

g1000 — Fri, 19 Jun 2015 22:07:13 GMT

You don't need to subscribe to Which, the report is freely available to download on the website.
There's no need to be rude, you first said ASA referred to download speed, then you changed your mind to sync speed, now you have changed it again to calculation of throughput from sync, so it seemed you were very interested in the matter and unsure about it. Sorry if you did not want to know direct from the source, then of course you don't need to ask.
That all ISPs use 76 Mb/s does not indicate to me they have no speed test evidence to back it up whatsoever. If they all have similar lines with identical sync, why wouldn't their top speed tests all be the same to a whole figure too. Probably they all copied one another ads too, safety in numbers.
I think your responses tell me all I need to know and that theres no point discussing further, have a good evening.