/ Technology

Win! Broadband providers are dropping exaggerated ‘up to’ speed claims

In a Which? campaign win, broadband companies are being forced to drop the unrealistic up to speeds they advertise following rule changes.

For too long, many consumers have been taking out broadband deals but getting speeds much slower than were advertised to them. But our latest research shows that’s now changing.

Previously, suppliers were able to advertise broadband deals which claimed ‘up to’ speeds that only one in 10 customers would ever reach.

But after our sustained campaigning, pressure from our supporters and rule changes from the Advertising Standards Agency (ASA) in May, some suppliers have now reduced advertised broadband speeds by as much as 41%.

New rules

The new advertising rules mean that at least half of customers must now be able to get an advertised average speed, even during peak times (8-10pm).

We’d been calling for these vital new guidelines to be introduced since 2013 through our Broadband Speed Guaranteed campaign.

We’ve now found that 11 out of 12 major suppliers have had to cut the advertised speed of some of their deals by an average 15%.

BT, EE, John Lewis Broadband, Plusnet, Sky, Zen Internet, Post Office, SSE, TalkTalk and Utility Warehouse previously advertised their standard (ADSL) broadband deals as ‘up to 17Mbps’. The new advertised speed is now more than a third lower at 10Mbps or 11Mbps.

TalkTalk has completely dropped advertising speed claims from most of its deals. While Vodafone has even changed the name of some of its deals: Fibre 38 and Fibre 76 are now Superfast 1 and Superfast 2.

And only Virgin Media’s advertised speeds have gone up since the change.

What’s next?

Alex Neill, our Managing Director of Home Services, said:

“Customers will now have a much clearer idea of the speeds that can be achieved when they are shopping around for broadband – even though their broadband won’t get any faster.

“With the change in advertising rules now showing the true landscape of broadband speeds, the Government must press ahead with its crucial plans to increase full-fibre availability and deliver the service that broadband customers need, without it costing them the earth.”

Have you bought a broadband package that hasn’t lived up to the speeds advertised? Have you noticed a change in how providers are advertising speeds?


This comment was removed at the request of the user

<i<Customers will now have a much clearer idea of the speeds that can be achieved “. No they won’t. All they know is a speed that 50% of subscribers should get and a speed that 50% will not. Which 50% are you in? How far down the speed list will you be?

The only speed that matters to me is the one I will get. so getting a personal speed estimate would be my priority.

This comment was removed at the request of the user

This comment was removed at the request of the user

This comment was removed at the request of the user

Here is a link to the data provided by Duncan: https://www.thinkbroadband.com/news/8132-uk-broadband-speed-test-results-for-july-2018 The Quality column – e.g. A (0.3) – is not explained.

Have a read-
I must confess I haven’t. Could you summarise it?

This comment was removed at the request of the user

Congratulations to Which? for helping bring us more honest advertising. I suggest that where the ‘average speed’ is shown in advertising that this should be accompanied by the recommendation that an estimate is obtained by the customer or prospective customer. When I was looking to move home in 2016 I obtained estimated speeds for properties I was interested in.

But not particularly useful! I don’t want to know what 50% of people get “on average”. Some will get more. But more important, 50% will get less than the average, some a lot less presumably. So how does this help them? And which 50% will I be in?

Because what I want to know (if I’m interested in speed) is the speed I am likely to get. This personal speed estimate is the only one of use to me.

We should be pursuing useful information.

A great deal more useful than ‘up to’ and if combined with a recommendation to obtain an estimate from the ISP and that’s the best that is likely to be achieved.

Seeing the data is useful.

For my PlusNet I have:

Plusnet ADSL/ADSL2+ A (0.6) Down 0.7 1.7 5.7 6.9 12.1 15.5

So, as someone who knows a little about how to use statistics,

This gives me 90% confidence that I’ll get 0.7 MBps or more;
80% confidence that I’ll get 1.7 MBps or more;
50% confidence that I’ll get 5.7 MBps or more;
20% confidence that I’ll get 12.1 MBps or more;
10% confidence that I’ll get 15.5 MBps or more.

Also the difference between the median and the mean gives some indication of the skew of the distribution.

For risk-averse purchasers (e.g. “glass half-empty types”), looking at the 10% values would be a better bet than seeing the 50% ones.

The introduction says “Have you bought a broadband package that hasn’t lived up to the speeds advertised?”. Well, if you now think you’ll get the “speed advertised” then 50% of you will be disappointed. How does that help – other than trading off one fairly useless piece of information for another?

The only really useful recommendation is a personal speed recommendation. “Up to” was understandable by many I spoke to – unless you didn’t want to understand it. But not helpful except to show the upper limit.

This comment was removed at the request of the user

duncan, I agree, as I have in the past. I am really commenting on the lack of usefulness of the advertised speeds, and the enthusiasm, apparently, for a speed 50% won’t get. It is a difficult one to put numbers to, but I’d have thought it better to have a figure at your input to the premises than a “general” figure. But you have to advertise something.

Criticism of broadband speeds is generally based upon what people don’t get. So presumably they measure the speeds that people do get to achieve this.

When some criticise UK speeds I’d like to separate out the people who could get fibre but choose not to subscribe. If their speeds are included, not separated, in any critical analysis on the state of UK BB then it is misleading (but makes a better story of course).

There’s no fibre along our road so I do not “choose not to subscribe”, I am stuck with not being able to subscribe. The best I would get would be fibre for all the way except the final 200 yards or so, and then I assume only as far as the master socket. So, malcolm r, I will continue to criticise UK broadband speeds on non-availability grounds until I get fibre to my master socket. And by that I do not mean just to the pavement somewhere near my house.

Please insert “copper wire, and that” between “I assume” and “only as far” Sorry folks…

Hi Electronic Biker – If you are registered and logged in you have half and hour to edit your posts and find posts that you made earlier. I hope you don’t have to wait too long for decent broadband.

Well, now we’ve tackled broadband via the ASA by substituting one fairly meaningless figure for another we might more profitably direct our anger at Smart Meter advertising. If I heard correctly on the radio this evening (but I didn’t totally catch it) I believe it suggested saving £34.50 a year. It’s about time realistic claims were made for smart meters – at best, apparently, nearer £11 a year.

Which?, will you look at smart meter savings claims in adverts with the ASA? Or do you believe in what is suggested?

I’m never sure what “savings from smart meters” are supposed to relate to.

Am I supposed to notice that I’m currently using 167W and act accordingly or will my tariff be cheaper because no paid meter readers will be required?

Surely the thermometer determines our use of gas for heating, not a reading on a smart meter display unit. And what we have for dinner determines whether we use the oven and two, three or four hob rings. Compared with electricity, gas is by far the most critical energy consumption to moderate yet the practicalities for doing so are more limited [assuming that people are not going around half-naked in their homes in the middle of winter and have closed all the windows]. Again, so long as people use a certain amount of common sense they will be naturally economical with their use of electricity commensurate with their idea of an agreeable lifestyle and their means. I think switching to LED lamps has saved more for me than any amount of smart metering will do [assuming the cost of the new lamps is sunk and not factored into the running costs; I think the payback is quite short anyway].

A large number of houses must now have smart meters installed. Are there any real life examples of the savings actually being achieved as a direct result of only using the monitor and not using any common sense?

I have been strongly opposed to the roll-out of smart meters. I inherited them with my present house and since they were incompatible with my new energy supplier’s equipment, they had to be replaced with new ones. The industry was allowed to carry on installing the old SMETS1 meters, so there will be more expense when these are replaced – shared amongst all customers, whether or not they have smart meters.

By showing the cumulative cost of electricity and gas the smart display can remove the fear of the next bill for those who are struggling to make ends meet. My preference would have been for those who wanted smart meters to pay for them and to give them free to those in debt.

I’m not sure what this has to do with broadband speed claims.

I’m not sure what this has to do with broadband speed claims“. Claims!. We castigate through the ASA the BB industry for making claims some dislike, yet far more dubious claims are made for smart meters that falsely support the £12bn roll out at our direct cost. Perhaps we could make more use of our ire.

“……… old SMETS1 meters, so there will be more expense when these are replaced “

Is this untrue then?
“Richard Harrington MP, parliamentary under-secretary at BEIS, brought up the scheme in the House of Commons on Tuesday, after reference was made to MSE’s previous coverage of smart meters.

He said: “The SMETS1 and SMETS2 meters have been much discussed, and I can confirm that a software program is being developed that will allow full conversion between the two.

“That will be done remotely, so customers who have had the meters installed will not have to worry about people coming to their house and changing them again.”

I remember reading this information but have heard no more. What I do know is that earlier this year I had both smart meters replaced with new ones and was given a new smart meter display. Darren, who did the job, said that he had swapped many others.

Perhaps someone knows?

This comment was removed at the request of the user

Maybe Ofcom should have a look at how many people are looking at Which? Conversation on a regular basis. It’s probably not a major problem and therapy may be available.

This comment was removed at the request of the user

I must admit that once started I find it hard to put down if the conversation is rolling along and there are no urgent claims on my time.

This comment was removed at the request of the user

“Individuals with DSL access tend to sleep 25 minutes less than their counterparts without DSL internet. They are significantly less likely to sleep between seven and nine hours, the amount recommended by the scientific community, and are less likely to be satisfied with their sleep,”

Overall the cost to the German economy is 60bn $

Well 6.21 a.m. here so time to get up and go watering. But the last few hours have been fun on the computer ….. and I will aim to get a civilised siesta after lunch.

Another paragraph from the study says this:

The temptations to which individuals are susceptible vary according to age, the scholars find. Among teenagers and young adults (aged 13-30), there is a significant association between insufficient sleep and time spent on computer games or watching TV or videos in the evening, while for older adults (31-59) the correlation is with the use of PCs and smartphones.

The overall conclusion mirrors other findings in studies from as early as the ’50s, which suggests that distractions of any sort at bedtime when the following day has a fixed morning schedule can lead to an element of sleep deprivation. Interestingly I remember (but can’t find) one study which claimed the advent of the electric light in houses foresaw the end of civilisation, since people would simply not bother sleeping any more.

You are so right Ian, since I got my pc and have all the bridge games and news of the international tournaments, I do sleep an awful lot less. But then, I am 83 and very soon I shall be sleeping for far too much…..
All the best to you youngsters!

F-Lux helps a bit.

We switched to fibre today with the promise of minimum speed of a whopping 25 and if we are very lucky 35. The price we pay for living in the sticks I guess. Anyway I rang our provider and asked the very hard Q. If it’s meant to be minimum of 25 why are we only getting 17. The reply “ it needs 10 days to stabilise”. Is it me or is that a truly rubbish fob off?

This comment was removed at the request of the user

I was told the same when I first switched to broadband. In order to have an accurate measurement of speed it’s necessary to disconnect internal wiring and connect the computer by cable if you can. Internal wiring and WiFi connections can slow down measured speed.

“if its still below the advertised minimum”. I rather thought what was advertised was not a minimum, but an average speed that 50% of customers will get. So not knowing which 50% you fall into means you don’t know whether you will get the “advertised speed” or not. If you’ve had a “personal” minimum speed prediction then there would be a case, but this is the point about “advertised speed” – it really is fairly useless.

This comment was removed at the request of the user

I suspect that Laney refers to the estimate speed provided by the internet service provider rather than the advertised speed, which would now be a median speed related to the period of typical maximum demand on the system.

This comment was removed at the request of the user

Thanks duncan

Median does not change. It is a type of average and is explained on Wikipedia and elsewhere. In the case of broadband it would be calculated from actual speeds.

It’s fair to offer a discount or the opportunity to terminate a contract if the service does not meet advertised expectations, which would be a speed estimate for the premises. As you have said before and I have mentioned above it’s essential to disconnect internal wiring and connect a single computer when testing speed. It would not be fair to the ISP if someone complained that their speed did not match expectations if all the family was online.

Roger Wyer says:
8 August 2018

7 miles from central London at St Margarets Twickenham we still dont have fibre optic connections!!

This comment was removed at the request of the user

Rocky says:
8 August 2018

This is a great result, however, my TalkTalk broadband issue is that it drops out several times a day. Until recently it meant manually resetting the WiFi router but I have since changed some router settings so that it now self-heals after an “intermittent sync” error. That’s an improvement but it still means frequent interruptions to my work of 3-4 minutes each time it drops out. Is this something to campaign about?

Have you discussed this with your ISP, Rocky? It’s a local problem, but the problem is to establish whether it is to do with your equipment, the phone line or interference affecting the WiFi signal.

My approach would be to unplug any extension wiring and plug a computer directly into the router. You could also borrow another router from a neighbour for test purposes. Some ISPs are happy to provide replacement routers free of charge, particularly where the existing one is old.

Rosemary Laidlaw says:
8 August 2018

Yes, I have a classic example of this issue that took me months to sort out with my provider
Last November their sales team rang me to try and persuade me to join their 5G scheme. As a Xmas present to myself and housebound I decided to sign up for a special rate of £35 per month for 18 months.with a new “router” able to deal with 5G which was switched on on the 22nd November. Absolute disaster as it would not perform in my area. I contacted them explaining but was told that I was out of the 14 day rule etc (I was bound to be) and what followed was an absolute “battle royal” but at 75 years old I was not backing down because they were not providing me with what I signed up to.
May saw a complete turn around in attitude, they acknowledged I had a problem. Far too long winded to write here but at their suggestion my telephone line is now on “fibre” and the reception and speed phenomenal. To keep the price within £2 of my original deal I had to fore go overseas calls but with WhatsApp and Skype available no inconvenience was caused. Sorted with both sides happy.
My own “take” but could be wrong is that the system was not ready for all areas to receive 5G at the time of launch.

Congratulations Rosemary. I hope that we will all still be fighting injustice when we are 75.

Billious says:
8 August 2018

Unfortunately the providers have not reduced their prices to reflect the true speeds

Well done Which. Keep up your sterling good work on the important causes you champion

Gareth says:
8 August 2018

It really makes you wonder what kind of moron agreed to the previous guideline of only requiring 1 in 10 customers to be able to achieve a given speed before these companies were allowed to advertise it. How on earth did they ever think that was going to be sustainable? Stinks of a bribe to me.

ok the broadband speed lies have been put to bed hopefully. the next major problem is when is a contract not a contract ? when you start to change providers. every contract I have had has been changed by the provider within 3 months, a price increase, yes you can if you want change provider again with out a penalty, but you have taken a 12 or 18 month contract to avoid any hassle. solution, if a provider decides to alter your contract in any form they must pay you a penalty for the full term of your contract. I think you will find a contract will then mean a contract.