Improving Choice in Faceted Navigation
When it comes to decision-making, less is more. New research indicates that consumers are more likely to buy when their choices are limited. Meanwhile, e-commerce sites across the Web have adopted faceted navigation systems to improve online shopping. But faceted navigation may not simplify choice enough. The answer could start with better data entry.
It’s counterintuitive but by offering a wider array of options, vendors seem to reduce sales. Jared Spool’s recent interview of Barry Schwartz provides a nice overview, and Mr. Schwartz’s new book The Paradox of Choice goes into further detail. When the decision is complex, the key to higher sales may be breaking the decision making process into manageable pieces.
Many e-commerce sites provide faceted navigation systems to shoppers. These systems allow users to narrow a list of products, one aspect at a time. For example, an array of 151 digital cameras on CircuitCity.com is made less bewildering by a menu of features or facets: resolution, camera style, price range, and brand.
Facets on CircuitCity.com
With these, users can reduce the list of cameras to a manageable size. Facets are flexible and robust: because the navigation is generated from the products themselves, an empty result set is impossible. And facets don’t involve a hierarchical taxonomy, so users can filter in whatever order seems natural. I have written elsewhere on the importance of good browse tools; and like any good browse tool, faceted navigation does an excellent job of shifting the responsibility for domain knowledge away from the user and onto the content producer.
However, faceted navigation doesn’t solve the problem of choice overload. Circuit City’s five digital camera facets still present a total of 22 options. And, as Jared Spool is fond of pointing out, I’m probably not looking for a camera anyway: I’m looking for a way to take pictures. So while I may know what I’ll be photographing and how I’ll be displaying the results, I may not know how that translates into optical zoom and megapixels.
Some e-commerce sites provide product selection wizards that produce a narrow result set from a series of questions, i.e. “What size prints do you want to make?” This ought to be a perfect front end for complex decisions: at any given moment, the user is faced with one simple, understandable choice. In my experience, though, these tools are frustrating. They seem to ask the wrong questions and their results aren’t what they should be. My suspicion is that these wizards are special, separate projects for vendors, dissociated from the larger process of creating and maintaining facets. I contend that with a more integrated data entry workflow, vendors can increase both the number and the quantity of such wizards, making consumers’ choices more pleasant and manageable and increasing sales as a result.
I imagine the current facet-creation process involves the following:
- Enter name of facet
- Define possible values
Why not augment it with…
- Create a user-perspective question for the facet
- Map answers onto values
More time-consuming perhaps, but if this information is present for every facet in the system then the product selection wizard becomes a guided front end for the faceted navigation system. Every product category or sub-category automatically gains a wizard. And it needn’t be a special, separate tool: users can jump fluidly between this Q&A filtering and traditional navigation.
Naturally there are pitfalls. While the employees who create facets presumably do have some domain knowledge, they may not be accustomed to creating user-oriented questions (”What size prints do you want to make?” vs. “How many megapixels do you want?”). But I imagine that with time, training, and supervision this will improve. And even questions that merely rephrase the facet should represent a marked improvement in the decision-making process itself, since the user will face only one simple choice at a time.