Understanding how consumers make decisions

Consumer researchers typically classify all possible options into one of three sets of alternatives: The consideration set, namely those cars that a consumer will seriously consider and evaluate; the inert set, namely the options that a consumer is tepid about (one could view these as back-up options); and the inept set, namely those cars that a consumer is unwilling to consider as viable options.

Marketing scholars have studied extensively the manner by which consumers decide which alternatives make it into their consideration sets. Let us suppose that in your case, you decide that you will only consider Japanese sports utility vehicles because you’ve had several past positive experiences owning Japanese cars. You eliminate premium brands such as Lexus, Infiniti, and Acura due to budgetary constraints. This leaves you with a pruned search space consisting of six models in your consideration set: Toyota, Nissan, Honda, Mazda, Subaru, and Mitsubishi. How will you choose between these six alternatives?

In their 1993 classic book “The Adaptive Decision Maker”, John W. Payne, James R. Bettman, and Eric J. Johnson explored how consumers choose which of many possible decision rules to deploy when making decisions (a meta-decision, since one is choosing how to choose). Take for example the lexicographic rule, which operates as follows: Choose the car that scores the highest on your most important attribute. Let us suppose that a consumer is evaluating the six Japanese cars in question on the following five attributes: price, maintenance cost, interior comfort, suspension quality, and engine power.

If maintenance cost is their most important attribute, and Nissan scores the highest on that attribute, it will be chosen if the lexicographic rule is used. On the other hand, the satisficing rule operates as follows: the consumer establishes a minimal acceptable score for each of the five attributes, and the alternative that is the first to pass all of these cut-offs is chosen. It is called satisficing because it might have been the case that all six cars would have passed all of the cut-offs, and yet the consumer will stop and choose the first alternative that passes all five cut-offs.

Let us presume that the attributes are scored on a common 1 (worst) to 7 (best) scale, the cut-offs for price, maintenance cost, interior comfort, suspension quality, and engine power might be 5, 5, 4, 4, and 3. This means that any car that is chosen must at the very least meet these minimal standards. If Mazda is the first car to be evaluated (e.g., because your mechanic friend highly recommended it), and it meets (or exceeds) all of the latter five cut-offs, it will be chosen (without ever examining the five other cars).

While there are many other decision rules to choose from, I’ll describe a third and final one, the weighted additive rule, which performs the following calculation for each of the six cars: Multiply each attribute score by its corresponding importance weight (as provided by the consumer) and add these to obtain an overall score. The car that scores the highest overall is the one that is chosen. This is known as the normative rule in that it is the one that a consumer should use if they wish to process all of the available information.

Let us suppose that the Honda attribute scores and importance weights (in parentheses) for price, maintenance cost, interior comfort, suspension quality, and engine power are as follows: 7 (0.40), 4 (0.25), 3 (0.20), 4 (0.10), and 2 (0.05). Note that the attribute weights add up to 1 to reflect their relative importance. The total score for Honda would be 2.8 + 1.00 + 0.60 + 0.40 + 0.10 = 4.90. The consumer repeats this calculation for the five remaining cars and chooses the one that has the highest overall score. For the sake of this hypothetical example, let us imagine that Honda yields the highest overall score, in which case it would be chosen. This implies that depending on which decision rule one arrives at one of three different winners: Nissan if the lexicographic rule is used; Mazda if the satisficing rule is used; and Honda if the weighted additive rule is deployed (incidentally my family chose the Toyota).

Choose the car that scores the highest on your most important attribute

How do consumers choose between these decision rules? This is precisely what Payne, Bettman, and Johnson explored in their brilliant book. They argued that a consumer engages in a trade-off between the cognitive cost of using a decision rule versus its likely accuracy. For example, the lexicographic rule is less effortful albeit less accurate than the weighted additive rule in that it does not process all of the available information (since it solely compares the alternatives on a consumer’s most important attribute).

A consumer who is choosing between competing toothpastes is likely to use a less effortful decision rule such as the lexicographic rule because the costs of making a poor choice are not very consequential (accuracy is not crucially important). Alternatively, if an individual is choosing between competing suitors to marry then it might be best to avoid the lexicographic rule and accordingly deploy the weighted additive rule.

There are important practical implications in understanding the decision rules that consumers are most likely to use within a given product category. In many nondurable product categories, a great majority of consumers will utilise the lexicographic rule using the same most important attribute (e.g., using price as the key determinant when choosing between toothpastes). In such situations, if your product does not score the best on the most important attribute, your product will never be chosen.

Even though you might have the overall best product when all of the attributes and their weights are used in the evaluation, this is immaterial because consumers are not using the weighted additive rule. I have used this exact insight in explaining how perfectly rational people might have voted for Donald Trump in the 2016 US elections (see The Parasitic Mind: How Infectious Ideas Are Killing Common Sense, pp. 31–32). Suppose a voter solely cared about immigration policy and he/she perceived Trump as scoring the best on that attribute (i.e., the lexicographic rule), the vote would have been cast in Trump’s favor (even though the voter might have chosen the competing candidate if another decision rule were used). The bottom line is that the same consumer might arrive at completely different choices as a function of which decision rules they deploy in a given situation. Understanding these cognitive processes is an important arsenal in understanding your consumers.

Dr. Gad Saad is a professor of marketing at Concordia University, evolutionary behavioural scientist, and author.