Virgin Wines “Pre-mixed case” multivariate test delivers full-bodied growth
Virgin Wines is an independently owned and run online wine retailer licensed to use the Virgin brand. Its range comprises of over 500 boutique wines at any one time, with 90% exclusive to Virgin Wines.
Founded in 2000 as part of the world famous Virgin group, Virgin Wines is the UK’s #1 online wine retailer. Selling over 80 million bottles of wine to over 1 million people. Working closely with winemakers around the world, using the millions of customer reviews to shape the style and taste of their wines. Virgin Wines were recently named as Online Drinks Retailer of the Year beating off the supermarkets and established retailers at the Drinks Retailing Awards – the Oscars of the drinks industry.
Virgin Wines has a reputation of delivering great wines from across the World. Their loyal customers are always looking to experiment with new wine varieties and one way to do this is to sign up for a preselected mixed case, where Virgin select an assortment of bottles and bundle them into a special priced offer. Whether it’s the summer barbeque, the Christmas season or any other celebration, these mixed cases never hang around for long. Not only do they represent excellent value but are often the springboard for subsequent single variety case purchases.
As part of an ongoing series of testing and optimisation, there was an opportunity to test the mixed case web pages to see if there was any room for improvement for both the visitor experience and see if there was any potential to increase sales.
“This test has strengthened the business’ belief in a programmatic approach to testing, targeting and web page optimisation.”
Head of E-Commerce, Virgin Wines
In this particular multivariate test, there was a hypothesis that visitors landing on these pages would find the information difficult to read (especially on mobile devices) and therefore the format could be tweaked to provide more precise information which would scan better. Factor one was to change the descriptive text from a paragraph promoting the bundled offering to more succinct bullet points about the wines. Factor two was to show the reduced bundled price as a percentage off the normal price, with a further variation showing the monetary value saved. The control showed the original price and revised price.
Both tests were carried out across desktop, tablet and mobile devices over a month-long period.
“This once again proves that what we think may work, doesn’t always, and how minor changes, such as showing the value as a percentage, can make such a dramatic difference to our bottom line.”
Head of E-Commerce, Virgin Wines
One of the main benefits of testing is to both prove, and disprove “gut feeling” that is often the only method that marketers use to improve the performance of their website pages.
Within this multivariate test the results were, like the wines, mixed.
Factor one, where the descriptive text was replaced by bullet points, showed a negative uplift across all devices. It increased the number of clicks to the product details page but did not show any increase in conversions or add to basket.
Factor two, showing the discount as a percentage had a positive effect and seemed to stimulate purchase. There were almost 3% more clicks to “add to basket” and nearly a 2% increase in sales, amounting to a projected £113,000 additional revenue per quarter.
“The tests we have run since working with Webtrends Optimize have already produced some positive outcomes, but this test has strengthened the business’ belief in a programmatic approach to testing, targeting and web page optimisation,” said Stuart Brown, Head of E-commerce at Virgin Wines.