Published on 15-Sep-2011
Validated on 01 Mar 2013
"We knew IBM Product Recommendations would lift revenues. We just didn’t think it would provide this much lift. After the initial set up it runs by itself, constantly updating recommendations on our site based on new product introductions and changing customer interests. It’s substantially more hands off." - Senior Manager, E-commerce Merchandising & Analytics, The Orvis Company Inc.
The Orvis Company, Inc.
Smarter Commerce, Smarter Marketing
Savvy retailers are integrating the web with other channels to create a multichannel network that delivers a quality customer experience across the board. At The Orvis Company, for example, catalog and online sales are tightly linked.
• Drive additional sales and increase average order value; • Keep product recommendations relevant and up to date for the site’s 4,400 products; • Keep the cross sell effort manageable
• Implementing IBM Product Recommendations to engage customers with timely, behavior-based product recommendations • Recommending products that help in the selection process • Cross selling appropriate items
• Revenues from product recommendations increased by 70 percent • Recommendations on product details page as a percentage of total site sales doubled, rising from 3 percent to 6 percent • IBM Product Recommendations saved approximately 8 to 10 merchant hours each week
For more than 150 years the Orvis name has stood for outdoor traditions, quality, and customer satisfaction. Orvis is the longest-running mail order business in the United States. In addition, the company has more than 44 retail stores in the United States and United Kingdom and more than 500 dealers worldwide, selling the company’s world-famous fly-fishing gear, distinctive clothing, home furnishings, gifts and dog products. Orvis moved into the eCommerce space in 1998 with the launch of www.orvis.com. This award-winning website offers not only products but also content, including Orvis-endorsed lodges information, free fly fishing tips, and expert articles.
Savvy retailers are integrating the web with other channels to create a multichannel network that delivers a quality customer experience across the board. At The Orvis Company, for example, catalog and online sales are tightly linked. Many customers browse the company’s print catalogs, and then place their orders at orvis.com. In fact, the catalog is the most important marketing vehicle for online sales at Orvis.
Once those catalog shoppers go online, Orvis merchandisers want to ensure the best possible shopping experience for them. To do this, they build upon Orvis’ 150-plus-year history of customer satisfaction and quality products.
Providing useful product recommendations is a vital part of their effort to deliver a superior customer experience. As customers browse product pages or view items in the shopping cart, they are presented with other products to consider. Consequently, they quickly find the right product or combination of products.
Ratcheting up the level of automation
Until recently, Orvis merchandisers were struggling to keep product recommendations up to date for the site’s 4,400 products. To keep the cross sell effort manageable, they took the approach of populating best sellers at the subcategory level. They were concerned, however, about the time they had to spend keeping product pages current.
Merchants were also concerned that their approach was not delivering the optimal shopping experience. Moreover, they were “leaving money on the table” by not providing options that might drive additional sales and increase average order values.
To make recommendations more relevant and reduce the time spent on maintenance, merchandisers needed to ratchet up the level of automation. However, they were wary of fully automated solutions because most solutions take a black-box approach that does not allow them to ensure that recommendations are both relevant and appropriate. For example, behavioral data might show a high statistical significance with respect to men’s apparel being browsed or purchased in the same session as women’s apparel. That does not mean, however, that a man’s cardigan is a good cross sell recommendation for women’s blouses.
Making smart choices
Orvis tackled the challenge with IBM® Product Recommendations, which engages customers with timely, behavior-based product recommendations.
Merchandisers were particularly attracted by the solution’s flexibility, which supports the delivery of different recommendation styles that address customer needs at different points in the sales cycle.
On the product page, for example, customers are trying to decide on a specific item. Recommending products that help in the selection process is appropriate at this point. If the customer is looking at men’s hiking boots, alternative hiking boots or walking shoes are good choices for cross sell. When the customer is viewing the cart, however, complementary items make more sense. If the cart contains hiking boots, appropriate cross sell items might include socks, hats or gloves.
More importantly, Orvis merchandisers liked the strong controls that IBM Product Recommendations provides. They can manage all recommendations through business rules based on category, price, inventory, margin and other factors. An IBM implementation specialist consulted with the staff to discover key business-rule needs and share best practices learned as a result of implementing solutions for more than 100 clients.
Doubling cross sell revenues
The day IBM Product Recommendations went live, revenues from product recommendations increased by 70 percent, and the site has sustained that level over time. Product recommendations on the product details page as a percentage of total site sales doubled, rising from 3 percent to 6 percent.
In addition to the direct revenue lift, Orvis estimates that IBM Product Recommendations saves approximately 8 to 10 merchant hours each week.
Solution Component Software
- IBM® Product Recommendations
For more information
To learn more about IBM please contact your IBM marketing representative or IBM Business Partner, or visit the following website(s): ibm.com/software/marketing-solutions.
Products and services used
IBM products and services that were used in this case study.
IBM Product Recommendations
© Copyright IBM Corporation 2011 IBM CorporationSoftware GroupRoute 100Somers, NY 10589 U.S.A. August 2011All Rights Reserved IBM, the IBM logo, ibm.com, and Coremetrics are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate US registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml Other company, product and service names may be trademarks or service marks of others.