This One? That One? One of Those Other Ones? Oh, Never Mind: Managing Overabundant Choice
by Sheena Iyengar
Have you heard of the famous jam study? When I met with the head of Fidelity Research, he explained it as follows: “Consumers think having more choices is great, but they’re less likely to buy a jar of jam when they have to choose from a larger array. We offer our clients up to 4,500 mutual funds, so this study provides a mantra for us: Narrow it down!”
Then there was the McKinsey exec who said that because of an internal memo regarding this same study, consultants now practice the 3 × 3 rule, in which the client first chooses from among three choices, which can lead to another set of three choices, culminating in no more than a third set of three options. This rule of thumb for presenting people with choices is also used by bank brokers, personal shoppers, and Wall Streeters, proving the broad appeal and utility of putting a cap on how much choice customers have.
And such encounters with enthusiastic jam-study disciples weren’t limited to boardrooms and business meetings. Once, during a long flight, the woman seated next to me shared the details of some research she’d recently read about in the New York Times. Apparently, she told me, a few years back someone had done a study in a supermarket, using different flavors of jam. It turned out that when people were given a smaller number of flavors to choose from, they were actually more likely to buy a jar of jam than when they were given a wider variety of options.
I’m the one who conducted the jam study. With millions of books available to today’s readers, I believe it has clear implications for publishers. But don’t despair. I also believe there are ways to navigate the downsides of the proliferation of choice. As consumers, we can all take advantage of them. And as publishers challenged by new formats, the long tail, chunking, recombining, and more, you can benefit from understanding what helps avert the paralysis too many choices can cause.
Expertise and Other Aids
In learning how to negotiate the demands of choice, it seems to me that there are two important first steps. We have to change our attitudes toward choice, recognizing that it is not an unconditional good. And, when possible, we must increase our expertise to counteract the limits on our cognitive abilities and resources, enabling us to obtain the most benefit from our choices with the least effort.
Developing expertise carries costs of its own, though. We can become experts in some areas, like speaking a language or knowing our favorite foods, simply by living our lives, but in many other domains expertise requires considerable training and effort. What’s more, expertise is domain specific. It will carry over only imperfectly to related domains, and not at all to unrelated areas.
So what can we do when we want to choose well in an area in which we have no expertise? The obvious answer is to take advantage of the expertise of others. If you’re a choice provider, this means figuring out how to give effective assistance to inexperienced choosers without chasing experienced ones away.
Turning choice into a collaborative activity, one in which individuals rely on and interact with many other people, is another way to manage it. Take the Best Cellars chain of wine stores. In contrast to the typical wine store, with racks and shelves filled with thousands of bottles organized by the region of origin or the type of grapes used, Best Cellars offers just 100 different wines in its stores, with each wine preselected for quality and reasonable price. What’s more, the wines are divided into eight categories with self-explanatory names like “ fizzy,” “juicy,” and “sweet.” More detailed information about each wine is clearly displayed above the bottles, and the staff is willing to make recommendations in nontechnical terms. It’s not a store designed to cater to connoisseurs or those shopping for a special occasion, but for the average, everyday consumer, it’s hard to beat.
We can also take advantage of the wisdom of crowds to make choosing easier. The Zagat restaurant guide provides one example. Amazon, of course, provides another. One benefit of these and similar recommendation systems is that while they impose some order on a massive number of options, they don’t remove any options, so experts who want something that isn’t on the computer’s list of suggestions can still find it on their own.
Categorizing options can ease the burden of choosing. Reduce a choice set into a manageable number of categories, and within any category include a manageable number of alternatives. In doing so, you might not even feel as though you’re limiting yourself. To see this in action, I, along with two research assistants, Cassie Mogilner and Tamar Rudnick, staked out the magazine aisle in several Wegman’s supermarkets and found that shoppers actually felt that they had more choices if there were fewer options overall but more categories.
Arranging a smaller selection of magazines under a wide range of subheadings, like “Health & Fitness” or “Home & Garden,” created a structure that made choosing more efficient and more enjoyable. This ends up being a win-win because customers are happier with fewer options and magazine publishers save money on the cost of producing the extra options.
Categories can be as simple as a department store dividing its goods into, well, departments, or as in-depth as sorting a single type of product into different categories based on attributes that would otherwise be difficult for most consumers to recognize, like the flavor categories at Best Cellars. The crowd-driven form of categorization is perhaps best demonstrated by the use of keywords and “tags” on media-sharing Internet sites like YouTube and Flickr. Whatever form it takes, categorization allows novices to reproduce experts’ abilities to ignore the irrelevant options and focus their attention on the most promising ones.
Recommendations and categorization are both useful features for people trying to make a difficult decision. They make the decision easier by allowing us to borrow the knowledge of experts or crowds, and they help us develop our own expertise more rapidly than we would if we chose without assistance. Learning what others consider good and relevant provides us with a general overview of a given field, catalyzing our understanding of it and the development of our preferences within it. Becoming an expert in every domain of choice is impossible, but we can become experts in the process of choosing, learning how to use the expertise of others to improve our choices and our knowledge of choice.
Just as we can learn from others, we can also learn from ourselves. When making decisions based on multiple attributes, the way we approach the decision can significantly impact how well we can handle large numbers of options. Along with my colleague Jonathan Levav, Mark Heitmann of Christian Albrecht University in Kiel, Germany, and Andreas Herrmann of the University of St. Gallen in Switzerland, I conducted an experiment on the German Audi Web site, which allows car buyers to build their new cars to order, choosing everything from the engine to the rearview mirror from a list of options.
We compared two groups of Audi A4 buyers. One group first made their choices for the dimensions with the most options: interior and exterior color, which had 56 and 26 different options, respectively. From there they chose by number of options in descending order, ending with interior decor style and gearshift style, which had only four options apiece. The second group encountered the same choices in the opposite order, starting with categories that offered the fewest options and ending with categories that offered the most.
Although both groups ended up eventually seeing 144 total options across eight categories, the people who started high and ended low had a significantly harder time choosing. They began by carefully considering each option, but they quickly became tired and went with the default. In the end, they paid 1,500 euros more for their cars (some of the defaults were more expensive than other options) and were less happy with their choices.
The Role of Restrictions
Henri Poincaré, the celebrated French mathematician and philosopher of science, said, “Invention consists in avoiding the constructing of useless combinations and in constructing the useful combinations which are in infinite minority. To invent is to discern, to choose.”
I’d like to invert the second sentence and propose a corollary: To choose is to invent. What I mean by this is that choosing is a creative process, one through which we construct our environment, our lives, our selves. If we ask for more and more material for the construction—that is, more and more choice—we’re likely to end up with a lot of combinations that don’t do much for us or are far more complex than they need to be.
In a conversation with the master jazz musician and Pulitzer Prize–winning composer Wynton Marsalis, he told me, “You need to have some restrictions in jazz. Anyone can improvise with no restrictions, but that’s not jazz. Jazz always has some restrictions. Otherwise it might sound like noise.” The ability to improvise, he said, comes from fundamental knowledge, and this knowledge “limits the choices you can make and will make. Knowledge is always important where there’s a choice.”
The resulting action is based on informed intuition, or, as he calls it, “superthought.” In jazz, superthought goes beyond determining the “right” answer: It allows one to see new possibilities where others see only more of the same, and to construct the rare “useful combination.” Perhaps we can superthink our way through choice by learning the fundamentals of its composition, and then using this knowledge to create music where there might otherwise be only noise.
Sheena Iyengar is the S.T. Lee Professor of Business at Columbia University. This article is derived from her book The Art of Choosing, just out from Twelve. To learn more about the book, visit HachetteBookGroup.com. To reach the author, go to columbia.edu/~ss957.