Does Your Web Site Understand Customer-ese?


Share on LinkedIn

With more and more consumers doing more and more of their business online, companies need to be aware of the online customer experience. In my eight years in marketing, I have been to numerous web sites where the customer experience was inadequate, and I’ll bet you have, too.

Consumers need to have confidence in your company’s ability to deliver an online experience that is designed around their unique needs. Many companies, however, do not understand the needs of their customers and, as a result, lose sales opportunities—and customer loyalty and satisfaction.

One way to enhance your customers’ experience at your web site is through an integrated search. Companies are starting to realize that search can be a great sales and marketing tool, because search phrases contain an indication of the user’s intent and can be combined with other profile information culled from the online session to better target customers and provide them with personalized experiences.

Companies with large web sites and complex products must strive to understand the different ways their customers articulate their needs online and be able to deliver a rewarding interaction experience, or risk losing market share to competitors who do. But many companies are not able to successfully interact with their customers. A report published by Jupiter Research on search effectiveness (Online Self-service: The Slow Road to Search Effectiveness, by Zachary McGeary, Feb. 15, 2006) indicates that search output often lacks relevance and accuracy.

Yet, sophisticated intelligent search and analytics technology can enable companies to transition from a keyword-based search environment, where results are driven by pattern matching, to a natural language search environment, where responses are driven by a semantic understanding of search requests and content.


People articulate their needs in different ways and any attempt to analyze and interpret customer behavior based on keywords can be an unmanageable process. When, however, search queries are analyzed based on their semantics using natural language technology, they can be grouped into concepts or intents, making the process much more manageable. For example, “how to change my email address,” “modify personal profile” and “mailing address update” represent three questions based on the same intent.

Research (

Telco Intents

, March 2005) by InQuira shows that within the telecommunications industry, 83 intent categories cover almost 75 percent of customer inquiries. The ability to categorize search queries and manage the interaction experience on an intent-by-intent basis is a powerful tool for improving customer interactions and increasing conversion rates.

Through the use of intelligent search and analytics, companies can define cross-sell and up-sell opportunities and shorten the sales cycle. They can gather intelligence on what customers are searching for and as a result, create and publish relevant content that satisfies a particular intent.

Assume a fictional customer, Joe Smith, visits the web site of a large specialty retailer of consumer electronics to buy a digital camera. He searches for “digital cameras under $500.” With most search engines, the search output will be associated with keywords related to digital cameras but will not be as specific as providing information on digital cameras that cost less than $500. This is because many of the search tools used by enterprises are keyword based and don’t understand the meaning of “under $500.” They lack the ability to understand natural language and interpret in real-time the user’s intent for that interaction.

Imagine that the same retailer has implemented natural language technology. Smith searches for “digital cameras under $500” and lands on a visually stimulating results page with a list of all the digital cameras that sell for below $500, ordered by third-party value rankings. The search engine not only understands the concept of “under $500” but also infers that the user is probably value conscious within his price threshold. The results page lists five different models and includes promotions related to digital cameras: online discounts, accessories, a photo printer, memory cards or even a subscription to an online photo-publishing service. The output page meets Smith’s initial needs and anticipates his secondary needs. It is a rich and satisfying experience.

Figure 1

The company collects valuable information from user sessions. It can monitor best performing or under performing content and promotions, changes in conversion rates and revenue, and uncover new business trends.


Consider Chrysler’s web site,

, and someone searching on “PT Cruizer” or “PT Cruiser.” The intent-driven results page (Figure 1) that is returned combines graphical and text elements that enrich the online user experience.

The results respond to the customer’s initial inquiry on vehicle model search, even correcting for the misspelling. The top search result depicts an image of the PT Cruiser and offers pricing information. Customers can drill down for more details on the specific model or compare the PT Cruiser to other Chrysler vehicles. The Chrysler site anticipates additional customer needs and addresses them. For example, you can get a quote in real time or build and price a model based on your specifications and preferences. This call to action gives Chrysler the power to accelerate its online conversions.

The sidebar reflects additional custom content assets and promotions that are related to the specific intent. Chrysler promotes the PT Cruiser’s commendations and offers a money-saving plan aimed at engaging customers in the buying process.

The text results in the middle of the page provide relevant content generated from a natural-language search that leverages industry specific dictionaries and sophisticated linguistic tools.

Searches such as “how do I finance a new car,” “payoff my car” or “make car payments” are tied around the intent or the concept of financing and billing. The technology behind Chrysler’s web site is sophisticated enough to understand that visitors who search for these terms are normally interested in financing options. As you can see in Figure 2, the intent-driven response at the top addresses inquiries about bill payment and financing options.

Figure 2

Depending on the complexity of the application, implementing an integrated search can take six weeks to a few months. Companies must realize the potential of integrated search as a sales tool, define objectives for their web site and invest in a solution that is in alignment with their goals. By investing in the right technologies, companies can offer rich intent-driven customer experiences and enjoy higher conversion rates and proven ROI.

Tina Miteko
Tina Miteko joined InQuira in September 2005 as marketing manager, overseeing marketing strategy and programs. Miteko has extensive marketing experience in both software and hardware industries. Previously, she ran the marketing department of GarrettCom.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here