Persuasive Advertising - J. Scott Armstrong [28]
3) obviously uninformative categories (“Category A”, “Category B”). Satisfaction was assessed after the people drank the coffee (unbeknown to the participants, the same coffee was used in all cases). People who knew little about coffee were more satisfied with their choices when the options were categorized than when there were no categories. Surprisingly, these gains in satisfaction were almost as strong when the categories were obviously uninformative as when they were descriptive. The fact that uninformative categories were almost as useful as informative categories illustrates that they reduce perceived confusion for customers. The experiment provided evidence on short-term comparison; in the long term, advertisers can educate their customers about how to make rational choices, and informative categories can help in this effort.
I had suggested using about seven products per category. This is consistent with the study that I just described in which eight categories worked well. It is also consistent with Iyengar and Lepper’s (2000) experiment involving chocolates, where subjects who were offered six choices of chocolates were more satisfied than those offered 30 choices, and with their jam-tasting experiment, where customers were more satisfied with six choices than with 24.
Two experiments directly addressed the issue of the optimum number of options. One offered from two to 20 choices of pens in increments of two. As the number of alternatives increased from four to ten, purchases increased from 50 percent of the customers to 80 percent; however, as the choices increased to 18, the proportion of buyers decreased to 30 percent (Shah and Wolford 2007). In another study, 120 subjects were offered gift boxes that differed in color and shape, where the number of options varied from 5 to 10 to 15 to 30. Satisfaction with the buying process and the product peaked at ten options (Reutskaja and Hogarth 2009).
The use of a decoy was examined using an ad for The Economist. The offer was:
1. Internet-only subscription for $59
2. print-only subscription for $125
3. print-and-Internet subscription for $125.
Customers can see that choice #3 is clearly better than choice #2, thus distracting them from the hard choices between #1 and #2. They will take the easy choice, #3. Choice #2 is the decoy. When these ads were tested among 100 subjects, 84 chose the combination, while 16 chose Internet only, and none chose print. When the decoy (#2) was not provided, only 32 percent chose the combination while 68 percent chose the Internet subscription.The Economist apparently found the decoy approach to be worthwhile. Similar findings were obtained for selecting a travel destination (Rome vs. Paris) or a blind date (Ariely 2008, pp. 1–6).
With respect to the number of features and to the manner in which they are presented, a simulation was conducted of choices among 21 cameras given information on six features. Good choices can be made when some features are more important than others, and where the excluded features involved few trade-offs with the included features (Fasolo, McCelland, and Todd 2007).
In Germany, in the early 2000s, approximately 12 percent of the people participated in organ donations, while in neighboring Austria, nearly 100 percent did. Why was there a difference?
1.3.4. Make the recommended choice the default choice
To do nothing is in everyone’s power.
Samuel Johnson, mid-1700s
A default choice is one that will be used if the customer takes no action. People tend to “select” a default choice. They do not go out of their way to change the current situation. This principle is consistent with the law of inertia (the status quo bias).
Evidence on default choices
In their study on organ donations (mentioned in the lead-in to this principle), Johnson and Goldstein (2003) cited a 1993 Gallup survey showing that 85 percent of those surveyed in the United States approve of organ donations. However, approximately 45,000 people died