Ofcom on Thursday defended research comparing 4G and 3G network performance between U.K. mobile operators after EE threatened legal action over one of its findings.
EE’s network topped Ofcom’s results table in terms of average 4G download speed, notching up 18.6 Mbps to second-placed Vodafone’s 16.4 Mbps. It also ranked first in the upload league table with 17.6 Mbps, followed by O2 with 13.2 Mbps (see charts).
However, EE lags behind its rivals when it comes to 4G Web-browsing speed.
Ofcom found that across all four networks, 83.4% of Web-browsing test samples loaded a test Website in less than one second. While Vodafone, 3UK and O2 were all above average, EE came in at just 62.9%. Further, 29.9% of Web-browsing samples failed to load within 15 seconds on EE’s network.
EE is not happy about the findings of this particular test.
"We’re not saying Ofcom’s testing is inaccurate, we’re saying it does not reflect the customer’s experience of our network," a company insider told Total Telecom. "To suggest that nearly one in three Websites will not load on our network is just not true," he said.
"We’re in discussions to take legal action against Ofcom," the source added.
To measure Web-browsing speed, Ofcom used a testing application on an unlocked Samsung Galaxy Note 3 to record the time taken to load mKepler, a test Website developed by ETSI. 50 samples from each operator’s 3G and 4G networks were taken in every city that Ofcom studied.
According to an internal EE document seen by Total Telecom, the operator has carried out tests of its own. It compared the same software used in the Ofcom test with the native Safari Web browser on an iPhone 5c, and also with Anite’s Nemo drive-test tool. It found that loading the mKepler test page took considerably longer using the Ofcom software.
When EE attempted to load the BBC Website, the speeds were a lot more consistent across t he three scenarios.
The tests were carried out in various locations at various times of day.
According to our EE insider, this suggests that the anomalies were caused by either an issue with Ofcom’s test software, the server hosting the mKepler Website, or the interoperability between that server and EE’s network.
"Ofcom’s research provides clear and comparable information on 3G and 4G performance across the main mobile networks," an Ofcom spokesperson told Total Telecom. "Operators that don’t perform as well as their competitors may well challenge the results. They would instead be better focusing on improving their performance."
They defended the use of mKepler in Ofcom’s experiment, pointing out that the content of Websites like the BBC changes on a daily basis, which has the potential to distort the results.
"What you see on the BBC in Edinburgh on Monday is probably different to what you would see in London on Friday," the spokesperson said. "The important thing is to treat each operator consistently."
To compile the report the U.K. regulator carried out 120,000 tests both indoor and outdoor across Edinburgh, Leeds, London, Newcastle and Poole/Bournemouth between October and December 2014. The tests measured download speed, upload speed, Web-browsing speed and latency.
"The report is intended to provide information that can help consumers understand how 4G and 3G mobile networks perform and support consumers in choosing a service that best suits their needs," said Ofcom, in a statement. "The findings are also expected to encourage providers to improve their performance."
