Amazon.com enables consumers to shop at the click of a button, which also provides rich information for manufacturers about consumer search and choice. One study shows how this information – which is freely and publicly available – can provide firms with insights into the competitiveness of their products.

Jun B. Kim, Paulo Albuquerque and Bart J. Bronnenberg use a statistical technique that analyzes the products that were searched together and infers how different brands and products compete in the market place. In addition, their approach allows the authors to visualize the competition in the market. Their analysis is unique in that they focus on what consumers search before they decide to buy rather than on purchase information.

“Our method visualizes consumer browsing behaviors in the form of product search maps. Manufacturers can use these maps to understand how consumers search for competing products before making a choice, including how information acquisition and product search are organized along brands, product attributes and price-related strategies.

“The maps also inform manufacturers about the competitive structure in their industry and the content of consumer consideration sets,” they said.

They used a statistical approach called asymmetric multidimensional scaling, which can show the relations of searched products in terms of price, brand, and other technical attributes for digital camcorders from Amazon.com for one year, starting June 2006. The data focuses on top four major brands (Sony, Canon, JVC and Panasonic) and the three most common media formats (DVD, hard drive and MiniDV), resulting in 62 products in their analysis.

An example of the information this approach yielded could be seen in MiniDV searches, which were found to cluster around products priced lower than $300. Manufacturers could use the layers of information from the map to identify their closest competitors – and even to see if their own products were competing against each other. For one Sony MiniDV, for instance, the top three most jointly searched alternatives were other, less expensive Sony products, while one JVC MiniDV was not searched frequently with other products, indicating it was not competing effectively.

At the category level, consumer searches tended to be guided first by format, second by price, and finally by brand. The method has a broader applicability and can be used to reveal broader patterns of search behaviors in other product categories.

“Manufacturers can use product search maps firstly to monitor in detail each product’s neighboring competitors during the consumer search stages. This would enable them to scrutinize the local relationships among products and thus to better understand substitution patterns for their own products during consumer search activities,” the authors said.

“Second, our analysis helps manufacturers to understand which and to what extent product attributes influence the contents of consumer product search. For example, we found that the role of brand is less significant than format or price, which is consistent with a recent large-scale industry survey.

“And third, they can use the product search map to diagnose the performance of their product lines.

“As more online data become available to practitioners and the marketing research community, well-tailored multidimensional scaling techniques may prove to be useful exploratory tools for analyzing and enhancing our understanding of brand search and consideration.”