June 26, 2009, 12:00 AM

How Improvement Direct improved average order value with A/B tests

After conducting A/B tests related to its online payment methods, the web-only retailer of household products increased revenue per visitor and average order value. A separate test revealed it could charge a restocking fee without hurting sales.


After conducting A/B tests related to its online payment methods, Improvement Direct Inc., a multiple web site retailer of household products, increased revenue per visitor by 33% and average order value by 38%, vice president of marketing Brandon Proctor says.

A separate A/B test revealed that the retailer could charge a 15% restocking fee without having either a positive or negative effect on sales, he adds.

Improvement Direct conducted the tests with Omniture Inc.’s Omniture Test&Target; tool along with the vendor’s SiteCatalyst web analytics application.

The test on payment methods-which offered various payment options to random groups of shoppers-revealed that the retailer may have overwhelmed some shoppers with too many payment options, and that removing some options led to more customer spending, Proctor says.

The retailer didn’t say which payment methods it removed, but the removed methods did not include credit cards, Proctor says. He adds that the removed methods appeared to be having problems processing payments, at times serving up error pages that apparently caused shoppers to either choose another payment method or just abandon the web site, he adds.

The payment methods still offered on Improvement Direct`s sites are the Discover, American Express, Visa and MasterCard payment cards; the online payment services PayPal and Google Checkout; eBillme, which enables consumers to pay for web purchases through online bill pay systems; and Bill Me Later, which lets consumers defer payment.

“Omniture Test&Target; showed us that removing particular payment options increased our sales metrics, an insight we wouldn’t have gained without multivariate testing,” Proctor says. He adds that the integration between Test&Target; and SiteCatalyst analytics provides for effective tests.

Proctor adds that the removed payment methods had been responsible for processing large amounts of payment volume, and that Improvement Direct will repeat some of the tests and dig more deeply into its analytics data to see how the problems might be fixed. If they can be fixed, the retailer will relaunch the payment methods to see if customers want to the use them again, Proctor says.

In the past, the retailer had absorbed the cost of restocking large returned items like bath tubs and armoires. After A/B tests showed that displaying a 15% restocking fee notice on its site did not affect sales on these large items, it now displays a restocking fee notice throughout its web sites and recoups what had been lost revenue tied to the costs of restocking returned products, Proctor says.

Improvement Direct is No. 127 in the Internet Retailer Top 500 Guide. Its e-commerce sites include FaucetDirect.com, Handlesets.com, KitchenDirect.com and ImprovementDirect.com.


comments powered by Disqus




From The IR Blog


Philip Masiello / E-Commerce

3 reasons retailers fall short in email and social marketing

Reason one: They’re constantly trying to sell their customer, rather than to help and engage ...


Rotem Gal / E-Commerce

7 surprising e-commerce trends for 2017

Consumers will engage with products and brands in new ways online in the year ahead.

Research Guides