ASA Ruling on Virgin Media Ltd
Virgin Media Ltd
Bartley Wood Business Park
30 October 2013
Internet (on own site), Press general
Number of complaints:
Summary of Council decision:
Two issues were investigated, both of which were Upheld.
a. Claims on the "Broadband" section of the Virgin Media website, www.virginmedia.com, stated "The faster your broadband speed, the more you'll be able to do online. So, if there are a few of you at home gaming, downloading, streaming movies and shopping, then mega speeds of up to 100Mb will let you all do your thing without slowing each other down. If you're going to do a bit less than that, but still want consistently fast broadband even at peak times, then you'll be fine with our up to 30Mb or up to 60Mb tariffs ... You get all this with your package too ... Our fastest wireless ... Free internet security ... Unlimited downloads Download and browse as much as you like with no caps and no hidden charges".
b. A press ad stated "Half price for 6 months ... £7.25 a month ... Broadband ... Unlimited downloads'1 - with no caps or hidden charges". Small print stated: "Broadband:1 Acceptable use policy applies. Heaviest users' download speeds may also be affected by traffic management measures from 4pm to 9pm and 10am to 3pm - see website for details".
1. British Sky Broadcasting (Sky), British Telecommunications plc (BT) and 20 public complainants challenged whether the "unlimited" claims in ads (a) and (b) were misleading and could be substantiated.
2. Sky also challenged whether the claim "no caps" in ad (a) misleadingly implied that there were no provider-imposed restrictions on a customer's ability to download data.
CAP Code (Edition 12)
1. Virgin Media Ltd (Virgin Media) confirmed the ads were for their cable broadband services which were subject to a Traffic Management Policy (TMP). The terms of that TMP had changed since some of the complaints were submitted to the ASA in April and May 2013. Prior to May, the TMP applied a reduction of speed by 40% to those customers who exceeded the data download allowance. The current TMP reduced a customer's download or upload speed (depending on which stream had exceeded the data threshold) by 30% for 60 minutes. If the customer's data usage did not reduce in that period, the speed was further reduced by 10%, resulting in a total reduction of speed by 40%, for the next hour (thereby being traffic managed for a total of 120 minutes). Customers who did not reduce their download/upload activity during those initial 120 minutes would trigger that cycle of reduction until the end of peak hours (defined by Virgin Media as 4 pm till 11 pm weekdays and 11 am till 11 pm weekends) and would continue to be traffic managed in that way for a maximum of 120 minutes after the end of peak hours.
Virgin Media believed the effect of their TMP was not dissimilar to the slowdown generally experienced by internet users, and therefore their TMP was moderate, meaning the "unlimited" claim was acceptable. They believed that a consumer's likely expectations of a broadband service would be based on the advertised speed i.e. the speed claim made in the ad, rather than the lower speed they ultimately received.
In support of their claim, Virgin Media provided a copy of a press release by uSwitch, a free, impartial online and telephone comparison and switching service, which was based on research they had carried out. The research looked at 2.3 million line tests across the UK's 50 most densely populated towns and cities over a period of six months. Virgin Media believed the sample represented well over 10% of the UK consumer market and, through scale, was representative of the UK population. The tests were completed across different technologies including ADSL, fibre optic and cable, as well as unlimited and capped data products.
Virgin Media noted that according to the uSwitch press release, the general slowdown in broadband speed experienced by consumers, which was caused by the number of people accessing the network, at peak time (defined as 9pm) across broadband technologies was 28%. They noted further that in one city, the slowdown experienced was as much as 60%. They referred to Ofcom's "UK fixed-line broadband performance report" (Ofcom report) published in March 2013 which showed that Virgin Media's customers (with 'up to 30Mbps', 'up to 60Mbps' and 'up to 100Mbps' services) experienced only an average slowdown of 3%.
Virgin Media customers had been included in uSwitch's line tests and Virgin Media believed that removing them from the sample would lead to an underlying average peak time slowdown experienced of around 35%. In their view, that data showed the impact of their TMP on an individual (a reduction in speed of between 30% and 40%) was consistent with the slowdown generally experienced by consumers. Consequently, they believed the TMP was moderate and their "unlimited" claim had been substantiated and was not misleading.
Virgin Media noted that, in their report, Ofcom had specifically excluded "Market 1" consumers - typically the most rural areas of the UK - from the ISP package comparisons. They therefore considered that uSwitch's approach to look at the top 50 most densely populated cities and towns was similar to that taken by Ofcom. In addition, they said that Ofcom excluded customers who had a line length (that is distance from the exchange to the end user) of over 5 km whereas uSwitch's data had not. They explained that due to line attenuation, those longer lines could experience particularly slow speed problems.
Virgin Media also pointed to data in the Ofcom report which looked at the distribution of average peak-time speed as a proportion of maximum speed (defined by the Ofcom report as "the highest download speed that a broadband connection is capable of delivering") for ADSL2+ ISP packages (packages in which broadband is delivered down the telephone line). They believed that data demonstrated that up to 5% of customers saw a peak time reduction in speed of more than 30% with up to 40% of customers seeing a peak-time reduction of more than 10%. They said that circa 3% of their customers were affected by their TMP, a figure which they noted was similar to ADSL2+ customers who experienced a reduction in speed by 30%. Therefore, they believed the impact of their TMP was not dissimilar to that experienced generally by consumers.
Virgin Media believed a consumer's expectation of their broadband service, regardless of whether or not it was advertised as "unlimited", was that they would experience a slowdown in speed during peak hours. They said it was clear from the uSwitch data the average consumer (i.e. someone using the internet and not an average consumer of Virgin Media's services) based on their experience of using the internet would expect a general slowdown in speed at peak times and that slowdown could be as much as 60%. In their opinion, their TMP was well within consumers' expectation of an "unlimited" service, regardless of the technology used to supply it.
Virgin Media explained that even where their customers were under traffic management, they could continue to carry out typical activity at or close to their normal connection speed. They said that was relevant to the speed of the service they were using as opposed to the potential headline speed. For example, a cable customer under traffic management on their 30 Mb/s service would have more than sufficient bandwidth to complete the streaming of BBC iPlayer in HD at a reduced headline speed of between 18 Mb and 21 Mb (the resulting speed having been reduced by 30% to 40%). They explained the customer's normal connection speed was unaffected in respect of the activity and their usage was not hindered or prevented.
Virgin Media said, however, the impact of downloading a large file while under traffic management was the time taken to complete the task. Consumers were not prevented from downloading the file and the task did not become more difficult. They believed the additional time taken to download a large file while under traffic management would not be contrary to a consumer's expectations because of the general slowdown experienced during peak hours.
2. Virgin Media said it was common within the telecoms industry to refer to broadband and mobile services as "capped" or having "usage caps", "data download caps" or similar and that exceeding those caps meant consumers were charged or prevented from using that particular service.
Virgin Media confirmed their broadband services were not subject to any caps on the amount of data customers can download and they did not charge an additional fee once those thresholds were reached. Their customers could download as much as they liked without being subject to additional charges for use above a specified level.
Virgin Media believed the reference to "no caps" appeared in conjunction with their unlimited downloads claim with small print referring to the TMP. They believed that when the full claim was read in context, it was entirely clear as to what "no caps" referred. They did not believe the claim was misleading.
The ASA considered claims such as "Unlimited downloads Download and browse as much as you like ...", 'Unlimited monthly downloads ... We give you unlimited downloads so you can enjoy as many music tracks, films, photos and files as you want, without having to worry about going over your limit" and "Unlimited downloads" were strong claims about the advertised services and that consumers were likely to expect features of a service which were described as "unlimited" were not unduly restricted by provider-imposed limitations, and that where policies existed, which limited speed of access for example, the restrictions could reasonably be considered to be moderate only.
We noted Virgin Media's view that their TMP was consistent with the average general slowdown experienced by consumers when using the internet during peak hours, which was supported by the uSwitch report. We acknowledged it showed that some customers were experiencing a slowdown of as much as 60% and that the average slowdown was recorded as 28% and considered that, in principle, it showed consumers' experience of general slowdown in speed when using the internet at peak times. Consequently, we considered that type of evidence went towards demonstrating what a consumer's expectation of a broadband service would be and contributed to determining whether Virgin's TMP and others of a similar nature (i.e. reducing the download speed) were moderate.
While we considered the nature of the evidence held by Virgin Media was suitable, we had concerns about the methodology. We noted it carried out 2.53 million line tests over a six-month period across the top 50 most densely populated cities and towns and that Virgin Media believed that data also included IP addresses in rural areas and whose lines were longer than 5 km. However, we understood line speeds from each participating IP address were tested only once which we considered meant it could not allow for a comparison between line speeds at peak and off peak times for that IP address. In addition, we understood the line tests were carried out using the uSwitch website and that the sample was self-selecting. We were concerned that could have had the potential to result in the sample being biased towards higher speed customers, such as those who were interested in technology, for example, who would want to see how fast their service was, and towards those with lower speeds who wanted to see how poor their service was. There was nothing, however, in the press release which explained how that sample had been selected. Furthermore, it was unclear from the report whether the sample had been weighted in order to be nationally representative. While we understood that line tests were carried out throughout the day, because each IP address could only be tested once, we considered that it was not sufficient to help demonstrate what a consumer's experience of general slowdown in speed throughout the day was like. As such, we considered it could not support an "unlimited" claim for a service subject to that type of TMP.
The Ofcom report looked at, among others things, consumers' experience of speed in their broadband service by measuring maximum speed, average 24-hour speed and speed experienced during peak hours (defined as 8 pm till 10 pm). That data looked at speeds experienced by consumers with ADSL, ADSL2+, cable and Fibre to the Cabinet broadband services. It showed that slowdown from maximum speed to speed experienced during peak hours ranged from 6% to 16%. That report stated also that Virgin Media customers who received Virgin Media's up to 30 Mb, up to 60 Mb and up to 100 Mb services experienced an average slowdown during peak hours of 13%.
We noted Virgin Media's comments that the correct data to consider was the distribution of average peak-time speed as a proportion of advertised maximum speed for ADSL2+ products. However, we considered that it was appropriate to use the average actual consumer experience across all technologies (ADSL, ADSL2+, cable and fibre), which we considered reflected broadband use in the UK, as an objective measure on which to base consumers' likely expectation of the service. Although the headline speed is important, we considered that consumers were more likely to base their expectations of a service advertised as "unlimited" on how it performed in terms of variations in speeds during the day in actual use, rather than in comparison to the headline speed stated in the ad.
We noted Ofcom's data had been weighted to ensure it was representative of UK residential broadband customers as a whole. Data from 2009, for customers with an "up to 2Mbps" service was factored into the sample because sufficient sample sizes for that product were unavailable. That data had been collected from multi-thread download speed tests. Although we acknowledged that Ofcom's data was based on 2,000 participants over the month of November 2012, it had carried out 693 million tests, each participant was subject to 14,000 tests a day, those tests were carried out throughout the day and that it had been weighted to ensure it was representative of UK residential broadband customers. We therefore considered that data was more robust than uSwitch's and that it better represented customers' experience, and demonstrated a consumer's likely expectation of a broadband service.
As stated above, the Ofcom report showed a slowdown of maximum speed of between 6% and 16%, which equated to an average slowdown across all technologies of 10% and we therefore considered that a TMP which saw speed reduced by 30% (with the potential of a maximum 40% reduction) was not moderate. Ad (b) referred to the TMP which explained that "heaviest users' download speeds may also be affected by traffic management measures from 4pm to 9pm and 10am to 3pm". However, because we considered the restriction was not moderate, we considered the reference to the TMP in the small print was likely to contradict, rather than clarify, claims that the service was "unlimited". For those reasons, we concluded the "unlimited" claims in ads (a) and (b) had not been substantiated and were misleading.
On this point, ads (a) and (b) breached CAP Code (Edition 12) rules 3.1 (Misleading advertising), 3.7 (Substantiation) and 3.9 (Qualification).
As stated above in point 1, Virgin Media imposed a TMP which reduced a consumer's download speed by 30% with the potential of a further reduction of 10% (i.e. a total of 40%); a provider-imposed limitation which we considered was immoderate. The ad made specific reference to the high speed offered by Virgin Media's 100 Mb service and reference to the ability of a number of users carrying out activities like gaming, downloading, streaming movies and shopping, which implied that particular service was suitable and appropriate for that type of bandwidth activity. In addition, the claim "Unlimited downloads. Download and browse as much as you like with no caps ..." was likely to be seen by consumers to mean that there were no provider-imposed limitations on their ability to carry out those activities. Although there was not a ceiling on the amount of data Virgin Media customers could download, a provider-imposed limitation on their service nevertheless applied. Because that limitation affected their experience of using the "unlimited" service in a way that was contrary to their expectations, we concluded the claim "no caps" in the context of an "unlimited" claim was misleading.
On this point, ad (a) breached CAP Code (Edition 12) rule 3.1 (Misleading advertising).
The ads must not appear again in their current form. We told Virgin Media not to claim their services were "unlimited" and with "no caps" if they imposed restrictions that were more than moderate.