Ofcom finally publishes final broadband report
Tuesday 28 July 2009 00:23:59 by Sebastien Lahtinen

Ofcom has today published its final report on its broadband research encompassing both performance metrics carried out by SamKnows and consumer perception research and statistical analysis by GfK. Many had been expecting Ofcom to shy away from making direct references to named service providers as it had avoided doing so in the past, but it has surprised everyone, to an extent at least, by publishing 'ranges' of speeds, backed by an independent expert who has peer reviewed the research and given it a stamp of approval. They wouldn't expect ISPs who fare badly to try and pull it apart now would they?

The report which runs into over 100 pages (excluding a second 60-page annexe) focuses heavily on download speeds since it argues that there is a close correlation between download speeds and other performance indicators, and that download speed is the most common consideration (along with price) when comparing broadband packages. Disappointingly, it does not address the subjective support element which can often be important in ensuring a good quality broadband experience.

A key headline figure in the report is the average downstream speed of 4.1 Mbps which our own recent broadband speed test statistics support with a figure of 4Mbps average speed across one million tests over a far larger unique tester sample size. Ofcom's test results have been 'normalised' (a statistical process to ensure that samples reflect the 'real world') so a smaller sample is nevertheless statistically meaningful.

DSL vs. Cable

Probably the clear leader out of these results is Virgin Media, the main UK cable operator which offers services up to 50 Mbps. Unfortunately, it wasn't Virgin's flagship 50 meg service that made it to the headlines due to low take-up, but its entry level 10 meg service, which Ofcom reports outperformed DSL operators on 'up to 8 meg' services by double, achieving average speeds of 8 Mbps. The average speeds of DSL operators tend to be significantly slower than the 'headline speed', probably due to the line distance which causes DSL signals to deteriorate. Cable services do not suffer from this same problem (to a significant extent). On Virgin's 20 meg service, the average speed does drop to 13 Mbps, although this still compares well to the headline speed in context.

The only saving grace for DSL operators is the increased jitter (the difference in the latency; the time it takes for a piece of information to flow from the broadband user to a server on the Internet; or vice versa) and most so latency, which some gamers might object to, but as Ofcom suggests, this is going to go unnoticed by most users:

"DSL services offer significantly better performance than cable services on jitter. However, this is unlikely to have a significant effect on the user experience for most internet applications, although for some online games jitter is a very important measure as the stability of connection can be paramount."

Ofcom Report

We were particularly disappointed to see the lack of sample for Be/O2 in the Ofcom research which for April 2009 was only looking at their joint 8 meg service. This group is by far the most significant threat to Virgin Media from the DSL operators and thus some more research would have been appropriate.

Analysing the results for April, there are clear downstream speed reductions in the evenings during the peak times across both DSL and cable services, although the DSL speed decreases appear more significant, suggesting Virgin Media's cable network has much capacity spare.

Broadband Genie have quoted BT as questioning the appropriateness of the results considering they are in the process of switching from ADSL to ADSL2+ technology. Virgin Media of course have been moving customers to faster services for some time.

User awareness

The research outlined a number of areas where users were not as aware as they ought to be about their broadband connections. The report concludes that more than 60% of users were unable to identify their 'headline speed' (the "up to" speed the service is advertised at) correctly with an estimated 15% giving an incorrect figure. Users are generally aware that distance to the telephone exchange is a significant factor in how fast your broadband is (although there is no suggestion if this was DSL users or also included cable users for whom this doesn't apply), but very few are aware that the quality of the extension cabling in your house can have a significant effect on your broadband speed.

We would remind anyone on an ADSL connection - If your broadband is slow and you have telephone extensions in your house (all correctly fitted with filters, including for Sky receivers and alarm systems), talk to your service provider about whether you might benefit from installing an I-Plate - More information can be found on our video show you how to fit an I-Plate.

Some Conclusions

It is rather disappointing that Ofcom have only looked at the largest service providers only passing a brief comment in the body of the report that smaller providers exist. Whilst this does cover 90% of the current UK broadband user base, one of Ofcom's statutory duties under the Communications Act 2003 is "to further the interests of consumers in relevant markets, where appropriate by promoting competition". As such, focussing the attention of the media on the largest providers only seeks to make it more difficult for smaller providers to grow. We accept that it is not necessarily possible to compare providers of all sizes on an equal basis, but dismissing this market is a worrying trend from a regulator. It also allows some smaller operators who are running higher contention levels to escape scrutiny of being under a microscope as with their larger counterparts.

The Digital Britain Report was a sizeable report for journalists to read, but we wonder how clearly the message from this report will be conveyed

Lord Carter has been reported as criticising journalists who have reported on the Digital Britain Report without reading it in full, but considering the length of government reports and those from organisations such as Ofcom, one does wonder why this is surprising. We are pleased to see Ofcom working on consumer guides which are far more likely to be useful to the average user. The research clearly illustrates a lack of understanding by the consumer as to how broadband works, and this needs improving.

So in summary, the DSL operators (BT Wholesale and other LLU operators) have some catching up to do.

The fundamental issue with all things technology is education, and this is the challenge we intend to take up.