Mae'r cynnwys hwn ar gael yn Saesneg yn unig.

Voluntary Code of Practice on Broadband Speeds: mystery shopping research

23 Gorffennaf 2013

Overview

Background

1. The Voluntary Code of Practice on Broadband Speeds (‘the Code’), which was revised in July 2010 following previous mystery shopping in 2009 (-1-) and came into effect in July 2011 (-2-), requires internet service providers (“ISPs”) who are signatories to make clear and accurate information on the broadband services available to consumers at the point of sale. As part of Ofcom’s monitoring of compliance with the Code, we commissioned market research agency GfK to conduct mystery shopping which took place between March and May 2013 to assess compliance through ISPs’ sales processes, both online and over the phone. The full research report from GfK is attached as an Annex, and we outline below the key findings of the report and Ofcom’s planned next steps.

The principal duty of Ofcom, in carrying out its functions, is to further the interests of citizens in relation to communications matters and to further the interests of consumers in relevant markets, where appropriate by promoting competition. (-3-) In light of these duties, Ofcom has taken a number of initiatives to ensure that the way in which broadband is sold works in the interests of consumers. These initiatives are:

  • Introducing and updating the Code which ensures that consumers get clear information on the maximum speeds available on their particular line (and other speed-related information) with the aim to help them make more informed purchasing decisions.
  • Publishing regular research which allows consumers to compare fixed-line broadband performance across different providers, technologies, geographies and time of day. (-4-)
  • Publishing consumer guides to advise consumers on how to improve their broadband speeds and get the best deal on their broadband service. (-5-)
  • Providing a ‘UK broadband map ‘which allows consumers to see at a glance how their region compares across a range of metrics with other areas. (-6-)
  • Ensuring that there is a clear regulatory framework in place for the deployment of superfast broadband infrastructure. (-7-)
  • Discussions with the Advertising Standards Authority and their Advertising Code-making bodies CAP and BCAP on the clarity of broadband advertising. (-8-) The reviews have in our view led to significant improvements in how broadband is advertised, and we continue to have an interest in how ISPs advertise their broadband services to ensure that consumers are given clear, consistent and accurate information.

3. The broadband market has changed extensively since we undertook these measures. Speeds have increased significantly due to greater take up of superfast broadband, prices have fallen, and there is now a greater choice of speeds and packages. With the wide range of packages available, it remains important that ISPs provide sufficient information to allow consumers to make informed choices.

4. The Code, originally introduced in 2008, is designed to ensure consumers get the information on broadband speeds which they require in order to make informed purchasing decisions. Pursuant to its 7th principle, which requires Ofcom to monitor compliance with the Code, we first conducted a mystery shopping exercise in 2009 to assess whether ISPs who had signed up to the Code were complying with both the spirit and the letter of the Code. The research found that ISPs were not always doing so. In particular, we found that mystery shoppers were often given inaccurate information about the access line (i.e. maximum) speeds they could expect to receive on their line. As a result, a strengthened Code was introduced in 2010 which made a number of changes, including requiring ISPs to give consumers broadband speeds estimates in the form of a range rather than as a midpoint estimate where there was large variation in achievable speeds (most notably for ADSL broadband).

5. The revised Code was fully implemented by ISPs in July 2011 and we conducted a further mystery shopping exercise between December 2011 and January 2012 to assess compliance with it. This research was published on 15 May 2012. (-9-) The findings showed that the level of compliance by signatories to the revised Code had improved since the previous mystery shopping exercise in 2009 and better information was available to consumers through online sales channels, particularly ISPs’ own websites. However, information on speeds provided over phone sales channels was less complete than for online channels, particularly for some ISPs.

6. We conducted another mystery shopping exercise this year between March and May 2013 and the results are published in the Annex.

Key findings in 2013

7. Overall, we have found that the level of compliance with the revised Code has continued to improve since the previous mystery shopping exercise in 2011/12. For example, the results show an improvement in the provision of speed estimates over the phone without mystery shoppers needing to prompt for this information, and an increase in the provision of speed estimates in the form of a range on the phone (rather than a single figure estimate). Whilst there has been overall improvement and considerable progress by some ISPs, for a small number performance has fallen or stayed the same in some areas.

8. In 2011, Talk Talk (47%) and BT Total Broadband (48%) were significantly less likely to provide unprompted speeds information over the phone and Talk Talk (58%) (-10-) was least likely to provide a range. We discussed with both ISPs the changes which would be necessary in their practices and instructions to sales agents to ensure that any issues were addressed. As reflected in the results, we are satisfied that BT has implemented the required changes. We are also pleased that Talk Talk has showed considerable improvement in providing a range over the telephone, although its performance remains the same as in 2011/12 for providing this information without prompting.

Telephone calls

9. The research found that, as in 2011/12, most mystery shoppers received some sort of speed estimate over the phone (either in the form of a range or a point estimate, or both - see Figure 1 below). Overall, in 96% of phone enquiries made, an estimate was given to the mystery shopper in 2013 (93% in 2011/12). Among the large and medium ISPs, EE (-11-) increased the proportion of calls in which a speed estimate was provided (up from 91% to 100%). Karoo and Plusnet also provided speed estimates in all calls while all other large/medium ISPs provided speed estimates in at least 9 out of 10 calls.

Percentage of “shops” receiving a speed estimate on the phone (range or midpoint, prompted or unprompted)

10. The Code requires that the speed estimate is provided in the form of a range. The research found that 80% of mystery shopping enquiries resulted in a range of speeds being given over the phone (-12-), rather than a midpoint estimate (an increase of 12% since 2011/12). A significant increase was achieved across the groups of large and medium ISPs, although not every individual ISP increased on this measure. Mystery shoppers to Sky and Karoo (-13-) were provided with an estimate in the form of a range by a sales agent in at least 9 out of 10 calls. Since 2011/12, Talk Talk, as well as BT Total Broadband and Virgin (ADSL) have improved on providing estimates in the form of a range. O2 was least likely to provide a range, supplying this in 58% of calls. (-14-)

11. The 2nd Principle of the Code requires that ISPs must provide all consumers with information on their estimated access line speed as early as practicable within the sales process, and in any event before consumers are asked for a Migration Access Code (MAC) or personal financial details, regardless of whether this is conducted over the phone, in a retail shop or through the ISP's website. In 2011/12, we found weaknesses in phone sales processes for some ISPs, particularly in mystery shoppers not being given speeds information without explicitly requesting it. (-15-) In 2011/12, a speed estimate (either as a range or midpoint) was given without the need for prompting within 59% of all calls to ISPs. There has been an improvement in this area; now speed estimates are provided without prompting in 68% of all calls to ISPs. The greatest rise is among the group of large ISPs with 69% providing unprompted information, compared to 58% in 2011/12.

12. BT Infinity, BT Total Broadband, Sky and Virgin Media (ADSL) were the strongest performers in terms of providing a speed estimate without prompting in 2013; at least three-quarters of calls to each resulted in a speed estimate being given spontaneously. EE (-16-) along with BT Total Broadband, BT Infinity and Virgin Media (ADSL) improved on this measure in 2013, in comparison to 2011/12. Talk Talk (47%) and O2 (55%) were among the ISPs least likely to provide speed estimates without prompting. Karoo’s performance fell compared to 2011/12 (from 76% to 60%).

13. ISPs gave the following feedback on our findings for telephone calls:

  • Karoo has expressed its concern over the increase since last year in the number of mystery shoppers needing to prompt sales agents for further speeds information.. It said it had introduced several initiatives since last year to further improve its performance and following this year’s results, will take action to provide speeds information earlier in its sales calls.
  • Talk Talk said that it undertakes its own quality checks which show a higher level of compliance, with only a very small proportion of calls being recorded without a speed estimate being provided. Our findings showed that mystery shoppers had to prompt for information more frequently in calls with Talk Talk than in calls to other ISPs. Talk Talk indicated that its staff provide speed information before a sale agreement is concluded or financial details or MAC code requested. It may be that TalkTalk’s sales process means that this information is provided at a later stage than anticipated by the customer. (-17-) We have informed Talk Talk that it should provide information as early as practicable in the process to be fully compliant.
  • For O2 see paragraph 16 below.

Websites

14. We also tested the speeds information available on ISPs’ websites. As in 2011/12, all signatories to the Code, with the exception of two small providers, offered a speed checker which provided estimated access line speeds. As for enquiries over the phone, there was an improvement overall in the provision of speeds information provided to individual mystery shoppers via the websites. The percentage of assessments where any speed estimate was given via the website rose from 87% in 2011/12 to 92%. The groups of small and medium ISPs showed the greatest rise. All except O2 offered a speed estimate in at least 9 out of 10 mystery shopping website visits with EE and Plusnet providing estimates in all visits. Results for O2 fell since last year with 63% (down from 79%).

15. The provision of an estimate through the ISPs’ website in the form of a range was 80% in 2011/12 and remained stable at 79% in 2013 (-18-). The greatest rise was seen among small ISPs. For BT Total Broadband, Karoo, Sky, Talk Talk and Virgin Media (ADSL), a range was available in at least 9 out of 10 cases. Mystery shoppers were least likely to report that O2 and Plusnet provided speeds in a range format. The results for both fell from 2011/12. Plusnet’s estimates were reported in 65% of cases (down from 77%). Investigation of the data in Plusnet’s case found that the speed range estimate might not be displayed for the ADSL product in some cases where the shopper was also able to receive fibre broadband. As for O2, the mystery shopping exercise found that O2 provided range estimates in 6% of all cases (down from 65%). This is because a purchase of many of O2’s broadband products could not be completed on its website (with consumers being directed to telephone sales channels where O2 indicates a speed range should have been provided). O2 highlighted that where a purchase of the relevant broadband product could have been completed on its website, an estimated speed range was provided. The number of mystery shoppers who did visit this webpage where a sales transaction could be completed represents too small a sample (5 mystery shoppers) to produce statistically significant results.  However, all of these mystery shoppers were given a speed estimate in the form of a range.  

16.ISPs gave the following feedback on our findings for the website:

  • Plusnet has assured us that it has taken immediate steps to ensure ADSL customers will get a speed range estimate and will continue to review the sales process to ensure the required information is provided.
  • O2 has informed us that earlier in 2013 it made a decision to concentrate its acquisition activity through its call centres and had removed the option to complete a Broadband (or Broadband and HomePhone) order through its online channel for most of its Broadband packages. Therefore, mystery shoppers who visited one particular webpage were not given a speed range estimate online, though some received a mid-point estimate. They were directed to call its Telesales channel to progress an order. This is reflected in the results described above. O2 indicated that when the mystery shoppers called over the telephone, they should then be given the full speed estimate information in the form of a range. O2 also informed Ofcom that mystery shoppers who visited its online shop at an alternative web address at which a purchase of certain of its broadband products could be completed, would have received a speed range estimate. The number of mystery shoppers who did visit the speed checker on this webpage where a sales transaction could be completed represents too small a sample (5 mystery shoppers) to produce statistically significant results.  However, all of these mystery shoppers were given a speed estimate in the form of a range. This, too, is reflected in the results above. We have reminded O2 that, to be compliant with the Code’s principles, speeds information, including a range, should be made available as early as practicable within the sales process.

Other information

17. In terms of ISPs providing information other than speed estimates over the telephone, there was still a lower level of compliance although the provision of some this information has risen compared to 2011/12:

  • 40% of mystery shoppers were told, by phone, some information about the factors that can affect their broadband speed (up from 34% in 2011).
  • 17% of mystery shoppers were told that the actual speed may be lower than estimated (19% in 2011). Mystery shoppers to Sky (ADSL) were most likely to be provided with this advice (as in 2011/12).
  • 11% of mystery shoppers were explicitly (-19-) told about the Code (which includes those who were told about the ISP’s own policy as well as the Code) at some point during their phone call (up from 4% in 2011/12).

Next steps

18. We will follow up with ISPs where the provision of information has failed to improve or decreased to ensure they take the necessary actions to improve.

19. We will continue to conduct further mystery shopping in the future to monitor ISPs’ performance and to assess how ISPs with weaker areas of compliance exposed by the current research have addressed this.

20. We will undertake a review of the Code to ensure that its provisions are still appropriate for consumers’ needs in the fast changing broadband market, including discussing with ISPs as to whether the Code needs amending to take into account broadband packages which allow more accurate spot estimates to be provided, and therefore where it is less relevant to give an estimated speed in the form of a range.

21. As choice continues to grow, consumers may need assistance in choosing the best package and service for them.  Six websites are currently accredited under Ofcom’s Price Comparison Accreditation Scheme covering price comparison calculators for fixed-line phone, broadband, and mobile and digital TV services. Cable.co.uk, Broadband Choices, SimplifyDigital and Broadband.co.uk are the Ofcom accredited websites covering broadband. In addition, some of these websites sell their accredited price comparison calculators to third parties as wholesale ‘white label’ services. Therefore a number of non-accredited price comparison websites use Ofcom accredited calculators. Ofcom is currently reviewing the scheme our consultation on proposals to amend the Scheme closed on 15 July 2013 (-20-).

22. We consider that the measures we have already taken help ensure that consumers have access to the information about broadband speeds required to make informed purchasing decisions, and we will continue to monitor and develop the following:

  • the Code, to include measures to ensure the Code remains appropriate for superfast broadband;
  • research and publication of information about the actual performance delivered by fixed and mobile broadband providers to consumers throughout the UK;
  • our accreditation of price comparison web sites; and
  • our consumer guides to choosing broadband services.

Footnotes:

  1.- See http://stakeholders.ofcom.org.uk/telecoms/codes-of-practice/broadband-speeds-cop/voluntary-code/.

  2.-http://stakeholders.ofcom.org.uk/telecoms/codes-of-practice/broadband-speeds-cop-2010/code-of-practice

  3.- Section 3 of the Communications Act 2003.

  4.- See http://stakeholders.ofcom.org.uk/market-data-research/other/telecoms-research/broadband-speeds/broadband-speeds-nov2012 for the most recent fixed-line research published in March 2013.

  5.- See http://consumers.ofcom.org.uk/2010/07/how-to-improve-your-broadband-speeds

  6.- See http://consumers.ofcom.org.uk/2011/07/first-interactive-uk-fixed-line-broadband-map-launched

  7.- See http://media.ofcom.org.uk/2013/04/25/uk-broadband-competition-reaches-new-milestone/ and http://stakeholders.ofcom.org.uk/binaries/consultations/fixed-access-market-reviews/summary/fixed-access-markets.pdf

  8.-http://bcap.org.uk/Media-Centre/2011/Changes-in-advertising-of-'unlimited'-and-broadband-speed-claims.aspx.

  9.-http://stakeholders.ofcom.org.uk/market-data-research/other/telecoms-research/broadband-speeds/bb-speeds-mystery-shop-12/

  10.- In the 2011/12 report, this data was reported based on all calls where any speed was provided, rather than all calls. The data reported above is based on all calls, in line with the approach taken in this year’s mystery shopping exercise.

  11.- Formerly Orange - assessed under Orange in the 2011/12 mystery shopping exercise

  12.- In the 2011/12 report, this data was reported based on all calls where any speed was provided, rather than all calls. The data reported above is based on all calls, in line with the approach taken in this year’s mystery shopping exercise.

  13.- BT Infinity and Virgin Media (Cable) do not have to provide a range as the Code excludes this requirement for cable and fibre

  14.-In May 2013, O2 sold its home broadband consumer business to Sky. Our research was based on calls/website visits made prior to this change.

  15.- The mystery shoppers had been instructed by the research agency not to complete a sale but were to make general enquiries and end the call when the call was in their view coming to an end, e.g. if they understood that the next step was signing up for the service or giving their Migration Access Code (MAC). If they had not been given any speed estimate by this point, they were asked to prompt the sales agents for this information.

  16.- EE data is compared with Orange data collected in 2011.

  17.- the 2nd Code principle: Information at point of sale requires that ISPs must provide all consumers with information on their estimated access line speed as early as practicable within the sales process, and in any event before consumers are asked for a Migration Access Code (MAC) or personal financial details, regardless of whether this is conducted over the phone, in a retail shop or through the ISP's website.  

  18.- In the 2011/12 report, this data was reported based on all website visits where any speed was provided, rather than all website visits. The data reported above is based on all website visits, in line with the approach taken in this year’s mystery shopping exercise.

  19.- 16% of mystery shoppers were told about a Code of Practice or some other polices or codes at some point during their call

  20.-http://stakeholders.ofcom.org.uk/binaries/consultations/price-calculator-accreditation/summary/accreditation-scheme.pdf

The full document is available below