Real User Monitoring (RUM) Methodologies
The second collection method is through a framework for Web page timing, such as Web Episodes. Of the two methodologies, navtiming is preferred. Having said that, the navtiming methodology is not supported for every user. In particular, Apple's Safari browser on iOS and OS X did not support navtiming data collection in the fourth quarter of 2014 -- as was the case in the third quarter. Apple seemed set to reverse course with a beta release of Safari for OS X and iOS 8, which both supported the navtiming API. However, Apple walked this back with the release of iOS 8.1.1 which stated, "The Navigation Timing API has been disabled only on iOS due to performance issues." It's safe to assume the API will be reinstated once these performance kings are worked out. Beyond Apple, Google added support for the API on Android 4.0 Ice Cream Sandwich and future versions, while Microsoft has supported it since version 9 of Internet Explorer. For more information on data collection methodologies, head to the Third Quarter Situational Performance blog post.
Average Page Load Time Over Broadband Connections
After reviewing the average page load time measurements for broadband connections, Japan, Turkey, and Hong Kong had the fastest page load times with an average time of 2 seconds. In contrast, the slowest country was Brazil with an average load time of 6.6 seconds -- over three times slower than the leaders. Australia and Paraguay filled out the bottom three with an average load time of 5.4 seconds.
Average Page Load Time Over Mobile Connections
For mobile page load times, Spain topped the list with its speedy 850 ms load time (the only country to post a sub-second average page load time). Spain's gain is more than half a second faster than its third quarter 1462 ms time. The next fastest countries were Turkey and Iran, with their respective mobile load times of 1.2 and 1.6 seconds. On the low end, Taiwan, Brazil, and India showed the slowest average load times. Taiwan and Brazil had average load times of 9 seconds, while India reported an average of 7 seconds. Interestingly, these times are all higher than the ones measured in those countries in the third quarter. One potential explanation for this change could be a shift in the profile of content consumed within those countries - users browsed more content from sites with rich media, or pages grew heavier, possibly related to the holiday shopping season.
The report also compared average broadband page load times to those observed on mobile connections. When compared, we can determine the relative "mobile penalty." The mobile penalty indicates how much slower a page loads on an average mobile connection versus a broadband connection.
Of the 43 countries/regions recorded in the survey, 17 had a mobile penalty lower than 1.0x, meaning their average page load times were faster on mobile connections than on broadband connections. The lowest mobile penalty observed was Spain's 0.3x penalty, followed closely by Colombia, Turkey, Indonesia, and Iran, all of which posted 0.6x penalties. Hong Kong and Taiwan had the highest mobile penalties in the fourth quarter, with penalties of 2.0x and 2.4x, respectively. In short, pages in Hong Kong and Taiwan load twice as fast on a broadband connection than they do on a mobile connection.
Going forward, as more customers integrate the Real User Monitoring capabilities and more platforms offer support for the navigation timing API, the scope of the Situational Performance section may be expanded. For the entire picture,download the full Fourth Quarter, 2014 State of the Internet Report.