According to a new report from Radware, the time it takes for a retail website to load is getting worse, with load times now 47 percent slower than in 2012.
Analyzing the top 500 retail websites, Radware’s Ecommerce Page Speed and Web Performance State of the Union report found the median load time for ecommerce home pages has reached 10 seconds, more than three-seconds slower than 2012’s median load time of 6.8 seconds.
Median Load Times for Retail Home Pages Since 2012:
Radware also discovered that the median load time for the top 100 websites was seven percent slower than top 500. According to the company’s research, today’s median load time for the top 100 sites is 10.7 seconds, making them 62 percent slower than in 2012 when the median load time for the top 100 sites was 6.6 seconds.
Top 100 Retail Website Load Times Compared to Top 500:
Of the top 100 sites, only 12 percent had a load time under three seconds. Radware noted a site that takes more than three seconds to load generates 22 percent fewer page views and has a 50 percent higher bounce rate than a site that loads in one second or less.
Radware blamed retail website slow load times on images:
There are a number of reasons why web pages have slowed down, chief of which are images, which account for at least half of a typical page’s total size. According to the HTTP Archive, images are responsible for 61 percent of the page weigh for the top 100 sites. What’s worse, many images are not optimized to render quickly in the browser.
Of the sites Radware analyzed, 34 percent failed to properly compress images, and 76 percent did not take advantage of progressive image rendering.
Evaluating Time To Interact (TTI) Speeds
Radware’s study also evaluated each site’s Time To Interact (TTI), or the time it takes a page to display its primary interactive content. For the median ecommerce page, the TTI was 5.4 seconds, with nine percent of the top 100 pages taking 10 seconds or longer to become interactive.
The top 12 fastest retail websites based TTI numbers didn’t necessarily include quick load times, with one site (FineArtAmerica.com) having nearly a 25-second disparity between its TTI and page load time.
From the report:
We have provided the time to interact alongside each page’s full load time in order to give perspective into the distinction between the two metrics, and to illustrate that load time is not always the most meaningful measure of a site’s performance.
Top 12 Fastest Retail Websites Based on TTI Speeds:
The study claimed that the retail sites using a content delivery network (CDN) took a full second longer to become interactive than sites the sites not leveraging a CDN. Radware said the extra second could severely impact key metrics, resulting in a 3.5% decrease in conversions, 2.1% decrease in cart size, and 9.4% decrease in page views.
Compared to TTI numbers from last year, when Radware first began calculating the metric, the median page now takes 10 percent longer to become interactive than it did in the summer of 2013.
コメント