CAMBRIDGE, MA: Online retailers can boost conversion rates by delivering faster page load times and differences of even a few milliseconds can affect outcomes, new research suggests.
In the State of Online Retail Performance study, content delivery network service provider Akamai Technologies analyzed one month's worth of data from its customers' retail sites, equating to around 10bn user visits, looking at performance metrics from the perspectives of IT, business and user experience.
This found the optimal load time for conversions ranged from 1.8 seconds on desktop (12.8% conversions) to 2.7 seconds on mobile (3.3% conversions) and 1.9 seconds for tablets (7.2% conversions).
"Given the fact that retail pages are expected to deliver so much visible and invisible content – from videos and high-res images to third-party tags – this is an impressive achievement," said Tammy Everts of the sub-two-second load times.
"This does not mean that 1.8 seconds is 'fast enough'," she added. "It means that, as we continue to make pages faster, we should continue to see payback in terms of increased conversion rates (or whatever metric you care about)."
Slower load times adversely affected conversion rates in the study. A two-second addition to the optimal rate detailed above led to a 36.5% decrease in desktop conversions, to a 26.2% fall for mobile conversions and a 25.1% decline in tablet conversion.
But even a delay of 100 milliseconds in load times produced a measureable difference, reducing desktop conversions by 2.4%, and those on smartphones and tablets by 7.1% and 3.8% respectively.
The report also observed that slowdowns appeared to affect different device users in different ways. Thus, desktop users were more likely to react negatively in terms of conversions, while smartphone users were more sensitive to slowdowns in terms of bounce rate. Tablet users, meanwhile, emerged as the most patient of the three groups.
That may be because consumers are more likely to browse on mobile but actually convert on desktop. And, Everts noted, separate research has shown that tablet users tend own older models and so have learned to be patient.
In the State of Online Retail Performance study, content delivery network service provider Akamai Technologies analyzed one month's worth of data from its customers' retail sites, equating to around 10bn user visits, looking at performance metrics from the perspectives of IT, business and user experience.
This found the optimal load time for conversions ranged from 1.8 seconds on desktop (12.8% conversions) to 2.7 seconds on mobile (3.3% conversions) and 1.9 seconds for tablets (7.2% conversions).
"Given the fact that retail pages are expected to deliver so much visible and invisible content – from videos and high-res images to third-party tags – this is an impressive achievement," said Tammy Everts of the sub-two-second load times.
"This does not mean that 1.8 seconds is 'fast enough'," she added. "It means that, as we continue to make pages faster, we should continue to see payback in terms of increased conversion rates (or whatever metric you care about)."
Slower load times adversely affected conversion rates in the study. A two-second addition to the optimal rate detailed above led to a 36.5% decrease in desktop conversions, to a 26.2% fall for mobile conversions and a 25.1% decline in tablet conversion.
But even a delay of 100 milliseconds in load times produced a measureable difference, reducing desktop conversions by 2.4%, and those on smartphones and tablets by 7.1% and 3.8% respectively.
The report also observed that slowdowns appeared to affect different device users in different ways. Thus, desktop users were more likely to react negatively in terms of conversions, while smartphone users were more sensitive to slowdowns in terms of bounce rate. Tablet users, meanwhile, emerged as the most patient of the three groups.
That may be because consumers are more likely to browse on mobile but actually convert on desktop. And, Everts noted, separate research has shown that tablet users tend own older models and so have learned to be patient.
Data sourced from Akamai; additional content by WARC staff
