This website works best with JavaScript enabled. Learn how to enable JavaScript.

VWO Blog

A Blog Around Everything Conversion Optimization And Beyond

Never miss an update from VWO

A value for this field is required.

See What @Thomas_Pink_ Learnt When They Tested Their Homepage on 136k Website Visitors

The Company

Thomas Pink is a clothing retail business based out of London, UK. Their shirts are inspired from London’s Jermyn Street, home of traditional British shirt-making. Famed for being the authority on shirting, Thomas Pink slowly expanded their offering to the entire range of clothing for men and women. They have many physical store and also sell and ship their products worldwide on their website

To get more sales from their website, they decided to optimize the homepage. This was done in conjunction with the team at Practicology — an independent eCommerce consultancy with a global footprint.

In this case study we’ll be covering one of the A/B tests that they performed on the homepage which increased order completions by 12.18%.

The Hypothesis

To push more people through the sales funnel, Thomas Pink decided to test adding a shirt finder navigational tool. This, they hypothesized, would make it easier for people to find a shirt in the color and style of their choice.

Another change that they made, in the same test, was removing the content heavy middle section and thus move up the product images and links. This was done with the objective of simplifying the look and feel of the homepage.

Before I go to the details of the test, see how their homepage originally looked like:


The idea for this test came from a rigorous study of customer data from various sources like analytic data, consumer surveys and on-site questions. They also had many insights from studies of behavior who shopped from their physical stores. One of them being, that the visitors who engaged with the fitting room were more likely to turn into customers. To mirror this behavior online, plus to replicate the ability to ask salesmen for shirts in particular styles and color, they decided to test adding the shirt finder tool right on the homepage.

They also knew that a large part of their customer base was repeat purchasers. “So anything to aid them find a product quicker should help” said Lee from Practicology

This is how the new version of the homepage looked:


The Test

A total of 136,000 visitors became a part of the test and was ran for 30 days.
They created this test using VWO and tracked 3 things: number of order confirmations (the primary goal), revenue (and revenue/visitor) and engagement.

The result confirmed their hypothesis. Their visitors did find the shirt finder tool useful and it made the purchase process easier. To validate this with numbers, the variation recorded 12.18% increase in orders. The absolute revenue for the new version also increased by 11.6% with a growth of 14.2% in revenue/conversion.

The important learning for Thomas Pink from this test was that their customers cared about ease in purchase. And introducing a shirt finder tool right on the homepage made it easy for users to quickly get to their favorites. Also, the clutter free new homepage, Lee thought, had overarching benefits leading to incremental revenue.

Let’s Talk

This test was run in the month of Sep’14 and seems like Thomas Pink is already at their next test. I can see some new elements introduced on the homepage. Plus, you’ll be greeted with discount upto 60%, do buy yourself a new shirt! 🙂

Do tell me your thoughts about this case study in the comments section below. And don’t forget to tell me which shirt you bought @taruna2309 on Twitter.

Taruna is a Marketer at Visual Website Optimizer. She takes care of paid campaigns and is constantly on a search for interesting ways to promote VWO.

Comments (2)

Leave a Comment
  1. Was it possible to look at the 2 variables in isolation? I’m curious to know what lift the Shirt Finder alone provided. Thanks!

  2. Hi Rosanna,

    No, we cannot. That is why we advice businesses to make only one change per test. I initially thought of covering only the shirt finder tool in the study but that shouldn’t have been correct. That’s why I reported all changes that were part of this particular test.
    So, we can’t really attribute the increase in orders and revenue completely to the shirt finder tool.

    The best way to go about it is — form a strong hypothesis, change one business element (and not necessarily one website element) and test it against the original. I blogged about this some time back. You can read it here:

Leave Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes : <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Contact Us / Login