2. Create confidence through recommendations. Reducing options works well when the variations among products are relatively small. But for highly differentiated goods — books, prerecorded music and video, clothes, and many housewares — you can’t get away with offering a small selection. (Even if you could, you probably wouldn’t want to, because niche purchases can cumulatively contribute heavily to sales.) Instead, you have to offer a wide variety while helping consumers navigate the complexity so they still have a positive choosing experience. How do you give consumers enough confidence to overcome the complexity of a large choice set? By turning to the people who already have that confidence: experts.
Through study and practice, experts in any field learn to simplify, categorize, and prioritize information, and to recognize patterns. This allows them to create order out of seeming chaos. For example, a chess player thinking eight moves ahead is presented with as many possible games as there are stars in the galaxy. He or she can’t possibly consider every option. The critical difference between novice and master is the ability to quickly eliminate the vast majority of moves and concentrate only on the most promising ones. The novice suffers under the pressure of choice, but the master knows how to relieve that pressure.
In high-choice conditions, the ideal consumer is the most expert consumer. That doesn’t mean someone with in-depth expertise in any one type of product. Most people don’t need to become specialists in jam or mutual funds to make decisions expertly. In fact, even if they did become experts, their knowledge would be limited to a specific domain and would not allow them to make better overall choices. However, novice consumers can become expert general consumers by learning to rank and structure their choice sets the way that experts do.
Marketers can thus help novices make more educated guesses and create confidence in their choices by giving them easy access to expert reviews and recommendations. In other words, you can attract consumers by allowing them to skip over much of the information-processing component of choosing, thereby minimizing their cognitive stress and enabling them to make good choices. Even non-expert advice can prove useful when there is consensus among a large number of reviewers or when the consumer trusts the source. This is one reason for the popularity of shopping websites with user reviews (such as Amazon), and also for the growing popularity of retailers that post recommendations for some of the products they carry (such as Whole Foods or New York’s Fairway grocery chain).
Another way to give consumers access to recommendations, especially when tastes vary or ratings aren’t easily available, is to set up automated systems that generate suggestions based on consumers’ expressed preferences. These systems, also known as “electronic agents,” are software programs that guide people by analyzing their prior purchases or their answers to survey questions. If consumers are willing to invest a little time teaching a well-designed system about their preferences, then the system can serve as a personalized expert for them. People don’t usually trust programs as much as they trust other people, but trust in well-performing electronic agents tends to develop over time.
For example, the Internet radio service Pandora has acquired 50 million users who tune in for an average of 12 hours a month, even though (or perhaps because) they cannot directly choose what they’ll hear. Pandora’s “mission” is to “play only music you’ll love,” and it accomplishes this by combining human expertise with an automated system. First, trained analysts determine the musical attributes of every song in the database. (Pandora calls this the Music Genome Project.) Then, when users tell the system what music they like, it searches for other music with similar attributes. As they listen to their personalized music streams, users can let the system know how well it matched their preferences. Eventually, the system comes to “know” the users so well that they no longer have to provide feedback. They can just sit back and enjoy.