Because I am a bit (lot) nosey

[#1]

We see loads of problems with broadband lines here but I wondered what proportion of customers actually raise faults over a year or is that commercially sensitive

[#2]

I'd hazard a guess that its smaller than most users would predict?  The technology has moved on dramatically and many users are more 'savvy' than in the dark days of yore when getting things to just work was an art. 
But for those who do experience problems it can still take much longer than expected.  BT Openreach can still be quite inept if you get one of 'bodgejobsworth' types - getting much less of a problem though.  The current 'hotshot' in this area is a young female technician of about 25
Maurice

[#3]

I wonder whether the recent? burst of broadband faults are due to the changes from 20Cn to 21Cn and ADSL1 to ADSL2 and whether those changes effect you may depend on where you live.
For example, my Mum (91) lives in a town and is quite near to the exchange. She was changed to 21Cn and ADSL2 and has not noticed. It all just went smoothly and delivered a higher speed to her without any problems. (They even have fibre to the cabinet available!)
I live in a village some 3km from the exchange and was recently changed over. I get a faster speed when it works but interference causes huge problems and long periods when I am connected but the connection is useless - huge numbers of CRC errors. Its being investigated but BT are struggling. It is not an easy problem to solve.
I think BT got carried away with its technology and have implemented a system that is fine in a town (generally) but over longer lines does not have the resilience it should have. There is no compromise option that can give me some improvement in speeds over ADSL1 but also cope with the problems of long lines. Its all or nothing.
Yes, I reported the fault. Yes it is being worked on. But the percentage of people complaining will be low because the percentage living in villages is low.

[#4]

for ME  PN  is the  9th wonder  of the  world....
  i  want to moane!!!!!!  but  i  cant ,,    haaaa

[#5]

Hi all,
Like EnglishMohican, I have a long line, but I live in a town.  Initially I was connected via ADSL2, but got punished by DLM as the line had been very unstable.  After several visits from OR engineers and all possible solutions, which did not succeed, PN decided to connect me via 20CN as the DLM will not be very harsh.  Some you may remember my coming on to this forum in the past with Adam Walker owning the Isuue I had.  You have not heard from me since switch over to 20CN as I haven't had any cause to.
Recently, I was switched over to 21CN with ADSL1 (Max).  Switch over went smoothly and noticed an improvement in my speed as compared with what I got under 20CN.  Line is still very stable for last 230 hours (Nearly 10 days).
So, I would say well done PN (and, dare I say, BT). 
I would say, OldJim, you may find many PN customers like me.
Many Regards,
Chandu.

[#6]

I'd imagine that it might be seen as 'commercially sensitive' by some but it would be very interesting to know (and I suspect that the true answer would put Plusnet in a pretty good light, both for the actual figures and for their openness and honesty) so - how about it PN? 

[#7]

I don't think it would be commercially sensitive. We disclose it regularly during service reviews for our Partner and Corporate base.
I don't have the consumer numbers to hand, let me do some digging.
M

[#8]

So, thanks to our resident guru Lee, here's some numbers:
The fault rate is calculated by dividing the monthly fault volumes by the average working system size, multiplied by 365 and then divided by 28 or 35 depending on whether it’s a 4 or 5 week month.
20CN: 0.092
21CN: 0.09
This is current data from this weeks reporting.
The maths geniuses out there (Oldjim you can crunch the data) should be able to get the gist of that.
Volumes are very low as you can see.
Mark

[#9]

Jim,
Were you intending to stuff these figures up one of the troll professor973's orifices?

[#10]

Either I have misunderstood the numbers as that looks to be horrible
Translating how I understand it
Number of faults per month = N
Number of users = Y
Number of days in a month = Z
Hence total number of faults per day = N / Z
Number of faults per year = N / Z x 365 = F
Number of faults per user per year = F / Y
and you say that is 0.09 on 21CN
That means that on average 1 user in 11 will have a fault in a year unless you missed a % out somewhere

[#11]

Accepting Oldjim's translation as the correct interpretation, it surely suggests that a statistically average user will have one fault every 10 years - which does not sound so bad. This is especially so if you allow that a high percentage of faults will be caused by the user him/herself and be beyond Plusnet's control.
The problem really seems to me to be that this figure assumes an homogenous distribution of faults across all of Plusnet's customers. As I said further up the thread, I fear that some customers are more liable to BT type faults than others. People on long, overhead lines out in the sticks are much more likely to have a BT type failure than those who live in a town (there will be exceptions that prove this rule !).
The country cousin also receives a poorer service in the sense that even when the line is working, it will run much more slowly than many town based lines. This is likely to be an accepted consequence of living in idyllic country but could be considered a failure even though with BT's T&C it will not be a fault.
This figure is really not much more than a simple (management grade) kpi virtually designed to make things look "under control"

[#12]

I was working on the basis that with 500,000 customers they will be reporting about 45,000 faults per year.
Multiply that up for the other major ISP's and it is no wonder BT Openreach are struggling

[#13]

But you are assuming that all faults reported to Plusnet are then reported to BT Openreach. Surely the whole point of Plusnet's "use the test socket", "try a different router" is to eliminate those faults that are not BT Openreach's fault. I used to work in returns analysis and anything between 50% and 90% of returns were not genuine faults. My trade was totally different so that may not read across to Plusnet  but I'll still bet that 45000 is much too high.
The other side of the coin is just what provision Openreach make to handle faults - essentially, how many engineers do they have? They are only going to struggle if the resource is inadequate. Even if 45000 is correct and all the other ISPs are worse, BT is a big company with huge resources targeted at these faults.
45000 faults is roughly 200 faults per (working) day, assume that an engineer clears 7 faults per day - (one per hour, optimistic?) then we need 30 engineers for Plusnet faults. That's not many.
I am more interested in discovering whether there are more faults being reported since the adsl2 rollout and the ftc rollout started. The forums seem to have a fair few problems related to those two. Is it more than was originally forecast/planned for?

[#14]

Quote

I am more interested in discovering whether there are more faults being reported since the adsl2 rollout and the ftc rollout started

As per Mark's posts we can look at current fault rates but it might not be possible to look retrospectively at what they were like before those technologies became widespread.

[#15]

It is potentially futile drawing too much inference from the figures you can measure because there are all those that you can't.
For example, different users have different expectations and understanding about what performance they should and are actually getting. If my line trundled along at 1Mbps I'd raise a fault because I know it is capable of far higher (both theoretically and emperically) however someone else may have no idea that anything was wrong and thus wouldn't say a thing thus fall under your radar.
A case in point; I've found it quite fascinating looking at the speed test results for other people in my vicinity using USwitch's Streetstats service. My speed is significantly higher than the majority and regardless of the fact that the discrepency could be down to customer's wiring/equipment/etc - and nothing to the with the provider - it does make me wonder how many of them are aware that they could/should be getting far better connectivity.
Of course, Jim did state he was just wondering what proportion of customers raise faults and thus wasn't necessarilly going to infer anything in particular from that information e.g. as a measure of customer satisfaction of how 'good', for whatever definition of the term, the service provision actually was, or an indication of how many faults there actually are.
Mathew