Last month Ofcom promised to take tougher action after revealing that many UK ISPs were only meeting some aspects of its Code of Practice (CoP) on broadband speeds and failing in other areas (here). Today the regulator has released the full results of its study, including individual ISP data.

To address the shortfalls, Ofcom is going to (March 2010):

• Work with ISPs to agree a consistent and accurate way of calculating and presenting access line speed information and amend the Code accordingly;

• Amend the Code to require ISPs to commit to giving the access line speed estimate early in the sales process, i.e. before asking the customer for bank details or a MAC. Currently the Code only requires ISPs to give this information before completion of the sales process.

• Find ways of ensuring that ISPs give consumers better information on why and how actual broadband speeds may be lower than headline speeds.

• Explore with ISPs whether it would be appropriate to add a new provision in Code which allows customers to leave their contract period without penalty if the access line speed received in practice is significantly below the estimate given at the time of signing up.

The results make for disturbing reading and could be useful in assisting consumers to make their choice of provider. The report itself is chocked full of various graphs, statistics and different ways of interpreting data that will bore the pants off most people. Luckily some of the information is fairly self explanatory.


Ofcom's code requires, among other things, that ISPs provide consumers at the point of sale with an accurate estimate of the maximum speed that their line can support. However just 43% (average) of the big broadband providers informed telephone callers about their speed without first needing to be given a prompt by the caller, with BT coming out the worst on 28% and O2 doing the best with 54%.


Here we see a similar illustration to the first one, except this time focusing on the performance of website based speed checkers and estimates. It shows that 21% of fake shoppers failed to get an access line speed from the ISPs website. A further 8% were given the headline speed and 5% were given a range of speed.

Overall, a single point access line speed (that was not the headline speed) was provided to 66% of fake shoppers (Figure 4). This time around its Sky Broadband and Virgin Media UK's Virgin.net (Virgin National) division that stand out the most, with Sky failing to offer a speed to 30% of fake shoppers and Virgin giving the often misleading advertised (headline rate) to 35%.

Ofcoms Full Broadband Speed Code Study Results:
http://www.ofcom.org.uk/telecoms/ioi/copbb/voluntary_code_copbb/isp_compliance.pdf