I have recently seen a lot of articles that have discussed how e-commerce websites are performing slower than ever before and subsequently are failing to meet end-user expectations. Out of everything I have read, reports by Radware and LoadImpact really stood out; they made me think about the present state of e-commerce websites, and the potential implications a slow website can have on etailers.
Firstly, Radware have continued to provide extensive information about e-commerce page speed and performance in their ‘State of the Union’ white paper. I was originally introduced to their research when it was conducted by Strangeloop, and I’ve always found their outcomes insightful. As well as this, their infographics are great visual presentations of their material. One of their recent observations was that web pages have continued to slow down when compared to the same results from 2012. Rawdware state that the median page took 7.72 seconds to fully load, compared to 6.79 seconds last year. This is interesting to note; especially considering that the average online shopper expects web pages to load in 2 seconds, and that after 3 seconds, up to 40% will abandon the site altogether. With this in mind, are user experiences being hampered by slow sites?
Radware attempt to understand user experience by measuring the ‘time to interact’ (TTI). TTI is described as the point at which a page displays it primary interactive content (e.g. feature banners with functional “call to action” buttons). From the sites tested, the median TTI was 4.9 seconds. This again is nearly two second slower than the average expectancy. With this information, it could be claimed that end user expectations and experiences are not being met because of slow sites. This then had me thinking. If consumers are likely to leave sites that are not up to scratch, are the e-commerce site owners losing out on potential revenue and custom?
LoadImpact believes they are! In their ‘State of the Web’ report, 53% of survey respondents claimed they had lost revenue due to poor performance or stability of their websites. In a competitive market place such as e-commerce, companies can’t afford to be losing customers and money. LoadImpact also observes that e-tailers are engaging with activities to help monitor and predict response times of their websites. Many e-tailers now monitor application performance and response times , with over 30% claiming that they do some form of performance testing. Some may say I am biased, but I don’t think that 30% is enough.
With Christmas only four months away, e-tailers should have started to prepare for the rush. The winter period is always the key period in e-commerce and I am intrigued to see how websites are going to cope with the holiday load. I commend both Radware and LoadImpact for their reports and look forward to their next editions to see how websites faired over the Christmas period.