Writing the Book on e-Commerce

SAS® Analytics Test Effectiveness of Variety of Amazon.com Features

In less than a decade, Jeff Bezos has turned a business he operated out of his garage into one of the most recognizable brands on Earth. That business, Amazon.com, came of age with the Internet. It is ironic, however, that such a technology-driven business operating in a futuristic channel is renowned for its customer experience and overall satisfaction.

Customer Success Video

Check out this video to learn more about Amazon.com and its success with SAS.

View Video

• Windows Media Player
• Real Player

(Runtime: 4 mins, 55 secs)

---

Customer ViewPoint
You have questions; our customers have answers. Check out this video Q& A.

Diane N. Lye, PhD
Senior Manager, Data Mining, Amazon.com

View Video

• Windows Media Player
• Real Player

---

(Requires Windows Media Player 6.4.7 or higher or RealPlayer 6 or higher)

Once the world's largest bookstore, Amazon.com now offers the world's largest selection of just about everything. Whether you want to buy some new kitchenware or baby clothes; browse every conceivable title of new and used CDs, DVDs and books; or contribute a few bucks to your favorite presidential candidate, an online universe of commerce is a mouse-click away. Visit the site once, and when you return Amazon.com remembers your name, items you bought the last time you visited, items you browsed – all so that it can recommend other titles or products that might interest you the next time you shop.

"We work hard to refine our technology, which allows us to make recommendations that make shopping more convenient and enjoyable," explains Diane N. Lye, PhD, Amazon.com's senior manager for worldwide data mining. "SAS helps Amazon.com analyze the results of our ongoing efforts to improve personalization."

"As a beta site for SAS®9, Amazon.com gained access to improved analytic computing capabilities. For example, the multi-threading available in SAS®9 allows Amazon.com to apply multiple processors to complex analytic tasks. The ability to speed up that process is important to us," Lye says.

In addition, Amazon.com uses SAS to test the effectiveness of a variety of features, designs and technologies and to measure how well they served the customer. "While we don't use SAS for the operational implementation of these features, we count on SAS to enable our test-and-learn approach for evaluating new product page layouts and new search technology," Lye says.

Creating rewarding experiences
Amazon.com has one of the world's largest data warehouses that it uses to enhance personalization and other features for customers.

Whenever Amazon.com designs and implements a new algorithm – the mathematical formula that determines page and product presentations for each customer – the goal is to help customers find what they want and purchase it in a way that is simple and convenient. Lye's group uses SAS to analyze operational and financial data to determine the impact that test algorithms have on the customer experience.

"We do a lot of work that's focused on understanding whether ideas that different groups have developed actually have a positive impact on the customer experiences," Lye explains. "With SAS, we analyze all the data to determine whether different design enhancements improve the quality of the page and ultimately the overall customer experience."

Search yields smart analytics
Innovation analyses don't focus entirely on spending. Lye's group uses SAS to determine the most effective way to present information to the customer. "On a typical experiment, we might have 70 metrics," Lye says. "We can't perform that level of analysis on the back of an envelope or in a spreadsheet application. We need some pretty smart analytics, and SAS is the only place we've found that can give us that level of sophistication."

"With SAS, we're able to handle larger volumes of data and process them more effectively than is possible with anything else available in the marketplace," she adds.

SAS provides Amazon.com with the flexibility to write its own formulas for success. "We're constantly seeking to improve the methodologies that we use to evaluate our innovations," Lye says. "That means that we've moved a long way beyond the statistics that we learned in college. The ability to push the method envelope is very important to us."

Thanks in part to SAS, Amazon.com now has a way to measure the impact of innovation. "Using SAS, we don't roll out anything to the Web site that will not drive value for the customer," Lye says. "And simultaneously, when we roll something out, we know what it's going to do for the business and for the customer. That measurability is critical for us, and it's clear that SAS helps us do that."

Copyright © SAS Institute Inc. All Rights Reserved.