Direct marketers have long been familiar with the practice of testing, learning from those tests, and testing again -- and online retail doesn't have to be any different, says Linsly Donnelly. As Chief Operating Officer of www.Joann.com, the leading website serving millions of arts and crafts enthusiasts monthly, Linsly and the JoAnn.com team have always made testing a priority.
Here's how JoAnn.com ran multivariate tests using 12 different combinations of offers and creative to come up with an ultimate winning recipe (or combination of elements) -- one that increased average order value by 137%, revenue per visitor by 209%, and conversions by 30%.
While JoAnn.com already believed in the importance of running ongoing website tests, the team hadn't been able to test all the specific variations they wanted.
"If someone wanted to test a picture of one sewing machine against a picture of two sewing machines, we'd have said, 'That would be great, but we don't have time to do that,'" Linsly says.
JoAnn.com needed a way to test different creative concepts and increase sales simultaneously. They sometimes felt that, while it's intellectually easy to test, it's operationally impossible.
Linsly found that most CRM tools give analytic insights but don't support marketing and promotion implementation. Offermatica was the first vendor who could merge tracking with implementation. "We didn't have to upgrade our ecommerce system, and it was affordable."
Once Offermatica and its "plug-and-play" software were in place, the JoAnn.com team set up the first tests:
Different variations on art, featured products, and promotions. Our favorite? Free shipping versus "Buy 2 or more White machines and save 10%."
The promotion seemed like a no-brainer. "Who needs two sewing machines?" Linsly says. But the team was willing to test it.
Variations on featured products, 30% off all items, online only promotions, and free shipping on an entire order.
Cross-sell and upsell options.
The winning recipe on the sewing machine category page increased average order value by 137%, revenue per visitor by 209%, and conversions by 30%.
"The 'Buy 2 and save' actually beat the free shipping offer; people were pulling their friends together, and we sold enough sewing machines to outperform single purchases," Linsly says.
In fact, recipes including that element increased average order value by 74% and total sales by 110%.
While only one combination delivered the best total dollar conversion, the JoAnn.com team learned from the other permutations too. "Some had a higher click through rate, some had higher conversions. There was a richness of learning beyond just the winning recipe."
The cross-sell and upsell tests on the shopping cart page didn't perform as hoped. "We thought it would be great, but as we looked at the data, it wasn't outperforming our current cart ," Linsly says. "It was either right at the baseline or below it. The Offermatica team was watching it and pulled it."
JoAnn.com will continue testing as part of the 2005 business plan. Why? "There are a million reasons to test," Linsly says. Here are two highlighted:
Potential customers who visit brick-and-mortar stores or open a catalog convert at a rate of approximately 60%. But online, some of the top performers only hit an 8-10% conversion rate.
"Most of the basic stuff online retailers have already tried. Now, we need a better equation to increase conversion rates, and it may not be intuitive. Testing takes the risks out of trying the counter-intuitive ideas."
So many new buyers are flocking to online channels that the ratio of new customers to existing customers remains high. "In essence, new customers drive activity daily. We need to be able to continue to test, to make sure that what worked for the prior customers works for new customers, as well."