Shop Direct and Tesco have the fastest websites among Europe’s top retailers, according to a new report from Radware and Level 3.

According to data included in the white paper, 57% of online shoppers will abandon a site if it hasn’t loaded after three seconds.

So Radware conducted a survey of the IR400 European ecommerce sites to see if they met this expectation.

Each site was tested using a range of browsers but the findings in the report are for Google Chrome, which is the most widely adopted browser in the EU.

Overall, the median web page load time was 7.04 seconds, which is double the time it would take for a majority of consumers to abandon their purchase.

As mentioned the research found that UK retailers Shop Direct and Tesco had the fastest load times at 1.69 seconds and 1.72 seconds respectively.

This roughly tallies with a test I ran last year looking at the site speed of the top 10 UK retailers. It found that Tesco and Amazon had the fastest homepages, while Next clocked the slowest load time.

It’s interesting to note that Currys is among the slowest retailers in the Radware report, as it recently overhauled its website using responsive design. However the redesign clearly hasn’t yet had an impact on optimising the site speed.

The research found that the median load time for new visitors was 7.04 seconds, however around a quarter (24%) of sites took longer than 10 seconds to load and 8% took more than 15 seconds.

Interestingly, for returning visitors the median load time decreased significantly to 1.81 seconds.

The importance of site speed for ecommerce sites is well documented, and we’ve previously highlighted research which shows that a one second delay in loading time can cause a 7% loss in customer conversions.

Yet Radware’s research found that one in four ecommerce sites takes more than 10 seconds to load.

Resource requests

One way of reducing page load times is to limit the number of resource requests, such as CSS, Javascript and image files.

Each use of these resources represents one server round trip that is needed to pull all of the page’s resources to the user’s browser.

The median IR400 site had 76 resource requests, however one out of three of the pages tested contained 100 or more resource requests and 2% involved 200 or more.

This chart shows the number of sites vs. the amount of resource requests

David Moth

Published 19 March, 2013 by David Moth

David Moth is Editor and Head of Social at Econsultancy. You can follow him on Twitter or connect via LinkedIn

1719 more posts from this author

You might be interested in

Comments (5)

Andrew McGarry

Andrew McGarry, Managing Director at McGarry Fashion

Recoding 3rd party generic javascript code is a quick win many companies don't execute. And then compound the issue by throwing the kitchen sink at the user with far too many share buttons!

Thanks, Social Media Execs! ;)

over 5 years ago


Matt Berry, Online Search & Affiliate Marketing Executive at

Have anyone used Tesco's website recently?
Their homepage might be speedy to load, but trying to use the groceries shopping section is infuriating. It's horribly slow, prone to crashing and a usability nightmare.
The mobile version is almost unusable.

over 5 years ago

Malcolm Duckett

Malcolm Duckett, CEO at Magiq

This IS nice to see. We are often harangued by people who say Celebrus's (and Magiq's) ability to collect detailed behavioural data about individuals might slow a site down - so to see one of our customers (shop direct) leading the performance field is proof if it was ever needed that our Tag-Free Technology really does speed site's performance and remove the need for complex performance-dragging tag management solutions. Well done, Shop Direct!

over 5 years ago

Chris Wright

Chris Wright, Solution Strategist - Retail Practice at Demandware

Very bad methodology behind this:

1) Ignores Safari, ruling out anyone using Apple's native browser (e.g. lots of iPad, Mac users)

2) Not much data collected - only 3 tests over 3 weeks! Doesn't account for seasonality or outliers so cannot be said to be statistically relevant

3) Data collected from a Netherlands location meaning download times are useful for comparison at best and do not actually reflect real user experience!

4) No device segmentation - it's easy to build a fast site when users are on a PC on a broadband connection, this report would be much more interesting if it showed Mobile speeds too (Mobile is the fastest growing channel so ignoring it seems odd as well)

I wouldn't go round quoting headlines from this in a hurry!

over 5 years ago


Geoff Paddock

The report has highlighted a key issue. Sitemorse routinely tests page load times for websites across a variety of sectors, not just retail, and the results are published quarterly. See the summaries page for each of our surveys for data on hiow long pages took to load.

over 5 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.