Search Me: Online Search Shifts from a Navigational Tool to a Customer Service and Educational Tool

Financial institutions have offered online search tools for almost as long as they have had Web sites. To date, these tools have been offered, with varying levels of success, as a failure-based navigational tool, where users resort to search when they are unable to locate desired content through the menu navigation structure. Early iterations of search functionality have largely relied on keyword-based technology. With the rise of natural language search algorithms, used in search functionality by firms like MetLife (New York) and Nationwide (Columbus, Ohio), as well as client adoption of more accurate Web search technologies such as Google, online users are becoming more comfortable with search functionality and the prospect that it will efficiently resolve their problems.

In the past year, there has been chatter in various financial services industries regarding the viability of search as a primary navigation service in which users would choose to use search functionality first before they resort to using a tab-structured navigation scheme, in the same way that search is now the first place most users visit upon arriving at retail sites like Amazon.com. However, it is worth noting that this is far more of a challenge for financial institutions, particularly insurance carriers. Whereas a retailer such as Amazon.com has users who are trying to locate discrete and unique products from a huge inventory of merchandise, a financial services institution is selling products that involve significant configuration and require an understanding of how the relationship will proceed in the future. At Amazon.com, the relationship is merely a succession of largely unrelated purchases.

For this reason, search has taken a different turn in the financial services industry and is being used increasingly as both a customer service and educational tool, with the goal being to answer precise questions rather than to direct users to a specific product or area of the site. GEICO (Chevy Chase, Md.) is a good example of this shift, having recently added search fields on the top of every page in its online auto quote process. The tool is called "Have A Question?" and the underlying concept is that search technology can connect users with resolution in a more efficient manner than making them sift through a larger help page. Furthermore, while help content is generally deployed in a page-specific or context-sensitive environment (to reduce sifting), search allows a firm to answer questions that may not apply to the content on the user's current page but nonetheless act as obstacles to his or her continuing the task. Users type in a question and are returned a direct answer, not a link to an area of the site.

In theory, search can work wonders as a question-resolution tool. For pages that are utilized exclusively within one task flow (such as pages within an auto insurance quote), financial institutions can map out every logical question posed by a user in that task and make answers available by search while eliminating questions and answers that don't apply to that user task. However, the best lesson to be learned from the deployment of both keyword and natural language search in the navigational realm is that even great technology will fail to return relevant search results if the financial institution hasn't effectively organized its site content and knowledge base within the search tool.

For instance, with GEICO's online auto insurance quote, if a prospect types in the question "Do you offer discounts?" he or she is returned the answer "GEICO offers discounts on car insurance to members of the professional associations and organizations listed below. The discount is applicable in most states." However, the organizations themselves are not listed, no link to a discounts section of the site is offered and the customer may be left frustrated.

Ultimately, the success of these tools lies not only in the technology but in the organization of content tied to the tool and the ability to continually test, tweak and improve the question and answer sequencing of its search capabilities.

Tim Carpenter is an insurance industry analyst with Watchfire GómezPro in Waltham, Mass. He can be reached at [email protected].