The top 500 retail websites are 21% slower now than last year, which could impact brand perception and customer retention according to a new study conducted by Radware.
According to the study, the top 500 retail websites take nearly 10 seconds to load. What’s more, web pages were found to be 31% bigger and heavier. The report titled, “State of the Union: Ecommerce Page Speed & Web Performance, Winter 2014,” reveals that top retailer sites are still not meeting the demands of online shoppers, which over time can have double the negative financial impact as website outages.
Here are some key takeaways from the report:
The median page has slowed down by 21% in just one year: Top ecommerce pages now take 9.3 seconds to load, as compared to 7.7 seconds a year ago.
The top 100 sites are slower than the top 500: The top 100 retail sites are slower than the rest of the pack, insufficiently meeting the maximum threshold in which a typical internet user is willing to wait for a page to load, which is 10 seconds.
Pages are now bigger and heavier: The median ecommerce page is 31% bigger and heavier than just one year ago (based on resources - CSS files, images, etc. - and KB).
The adoption of core performance best practices has plateaued: While 80% of the top 100 ecommerce sites now use a content delivery network (CDN) (up 6% since spring 2013), keep-alives have plateaued at a 93% implementation rate, and image compression is still not widely adopted, standing at 9%.
Tammy Everts, web performance evangelist at Radware, said in a press release that 2013 had its share of website outages from Amazon to Healthcare.gov.
“We also see that site slowdowns can also cause a negative impact on brand perception,” she explained. “Slowdowns occur 10 times more frequently than outages, and over time, slowdowns can have double the negative financial impact as outages. This also has a major long-term impact on customer retention, as the permanent abandonment rate for a slow site is up to three times greater than the abandonment rate for a site that is down.”
Everts added that, at 5 seconds, the median time it takes to interact with a page does not meet consumer expectations.
“All of this equates to a longer wait time for the customer, who may abandon a page if it takes longer than three seconds to load,” she said.
The tests in this study were conducted using an online tool called WebPagetest – an open-source project primarily developed and supported by Google, which simulates page load times from a real user’s perspective using real browsers. Radware tested the home page of every site in the Alexa Retail 500 nine consecutive times. (The system clears the cache between tests.) The median test result for each home page was recorded and used in the calculations. The tests were conducted Jan. 16-26, 2014 via the WebPagetest.org server in Dulles, Va., using the latest version of Chrome (31.0) on a DSL connection.