Marketers and voice-of-the-customer (VoC) champions of the world, unite! You have nothing to lose but your shackles of outdated theories and measurements.
Despite the stir about behavioral economics turning traditional economics on its head, we all know that the first behavioral economists were marketers—the people charged with understanding and influencing the behavior of consumers. These are the people who said the product and market conditions are a given; now we just need to re-brand, re-position, re-package, re-message, re-target … re-whatever to get people to buy what we have.
Most approaches to measuring customer loyalty and experiences, understanding the levers or drivers of loyalty and satisfaction, and creating strategies and tactics for improving performance, however, remain largely encrusted in the outdated thinking of traditional economics. It’s time to cast off those chains and explore what behavioral economics means for measuring and managing customer loyalty and experience.
1. It’s Their Behavior, Stupid
The objective is to understand and motivate customer behavior. Loyalty as a mindset and the experiences of customers across whatever touch-points (or experience points) are important only to the extent that they motivate loyalty behaviors, those actions taken by customers—such as buying, buying more, buying more often, recommending that others buy, etc. – that determine customer lifetime value and create value for a company. Everything else is noise.
Marketers have to identify (and then validate) the research metric that best predicts the actual behavior they want from customers that creates the desired business outcomes. They then must determine the criteria against which they need to perform to earn the types of higher grades that will boost performance on key metrics and motivate loyalty behaviors.
2. Maximizing Variance Is Not the Same as Predicting Behavior
Researchers lose sight of the objective: understanding and then motivating behavior that creates value. Instead, they get lost in the effort to maximize statistical variance because most key driver models rely on regression-type approaches fueled by variance. These approaches get “better” models from maximizing variance. But this reasoning is circular and totally misses the point, as increased variance in the data may have nothing to do with customers or better understanding and predicting their behavior.
Getting variance is easy: elongate the scale, don’t use descriptive anchors on response categories, force respondents to interpolate between abstract numerical ratings, force rank orderings, perhaps add a dash of ambiguity in the questions and sprinkle in some mixed-mode surveys. Much of the variance in survey data is more noise from a “random walk” (I have shamelessly borrowed the term from Mack Turner at Bank of America) than a true measure of statistically meaningful differences in the underlying behavior being measured.
3. Intention Is Not the Same as Behavior
While perhaps obvious, “it’s a long step from saying to doing,” as Don Quixote and Sancho Panza reminded each other. Still, we see some marketers trafficking exclusively in stated intentions, while others are content to focus merely on the likelihood that customers will do something (recommend a brand, for example).
The research problem is compounded for those who zero in on stated importance as a critical measurement. As individuals, we tend to think we know why we do what we do, and we offer explanations for our decisions. But the motivations for human behavior are mainly immersed in the 90 percent of the brain that is invisible to us. Our explanations for our actions are mainly after-the-fact rationalizations.
4. Emo-rational Behavior
The economic model of people as rational decision makers is kaput. Traditional economics posited a world in which consumers are rational information processors who make optimal decisions to maximize their well-being. Behavioral economics challenges this paradigm on every point: People lack access to, interest in and the ability to understand all potentially relevant information. We make simplified, “satisficing” decisions; things people value are highly subjective and do not align with the economic concepts of utility, and people are absolutely lousy at assessing cost/benefit tradeoffs. Our rationality is, at best, bounded by an array of constraints, while our decisions are largely dominated by the 90 percent of the brain submerged in emotions.
The fact that we are not rational in the economic sense doesn’t imply that people are irrational, as this seems to indicate there is no logic or pattern to our behavior. Are our individual tastes and preferences, likes and dislikes, and desires and turn-offs irrational? Or are they simply the currency of our emotional lives? We value (attach utility in economics speak) stimuli that make us happy and try to avoid (attach disutility) things that cause displeasure. That actually seems perfectly reasonable, even when our actions violate otherwise rational guidance about avoiding risky behaviors and maximizing our health and wealth.
5. Memory Trumps Experience
People, as Kahneman politely puts it, “misremember.” Why can’t 10 observers of the same event agree about what happened? What we remember is a facsimile of the original experience. We are far more likely to remember the highs and lows of an experience, as well as the end point and those aspects with which we might specifically identify because of some prior memory, preference or interest.
When we make decisions as consumers—whether to buy again, buy more, switch providers, recommend (or not recommend) to others—we’re drawing upon the memory that has been imprinted, however perfect or imperfect a reflection it might be of the actual experience.
6. The Behavioral Mind: Bad Is Stronger Than Good
Cynics of the world, take solace: Negative information is recognized more quickly, processed more thoroughly, acted upon more readily, remembered for longer and passed on to others more often than positive information. Regardless of where one sits on the glass half full/half empty debate, people are hard-wired to accentuate the negative, not the positive. There is a reason why people share bad news and bad experiences more readily than good news/experiences.
This is why defense must be the brand’s first priority – because poor customer experiences will have a more forceful and lasting impact than great experiences on future customer behavior.
7. Behavioral Math: A Positive Plus a Negative Equals a Negative
A natural corollary of the “bad is stronger than good” axiom is that one dose of a negative experience plus one dose of a positive experience equals a negative impact. Good performance and experiences are not compensatory for lousy ones. Overcoming a bad taste from a problem or poor experience is a challenge, and it is costly.
Recovery from a problem or experience failure certainly is possible. Customers who have their issues quickly and thoroughly resolved (perhaps with the addition of some extra benefits) often remain loyal to the company involved. There are those who would even argue that people who have great problem resolution experiences can be more loyal than customers who never experienced a problem.
Failing to account for the greater impact of the negative is a major shortcoming of most approaches to analyzing and acting on the drivers of customer loyalty and satisfaction with their experiences.
8. Differentiate Between Upside Opportunities and Downside Risks
Not only is bad stronger than good, but the drivers of bad usually are not the same as the drivers of good. For any product or service, there are basics that customers assume and take for granted—throw the switch, the lights will go on; the hotel room will be clean; the coffee will be hot. Much to their chagrin, Con Edison, Marriott and Starbucks, however, earn zero points when the lights work, the rooms are clean and the coffee is hot, as these are the baseline expectations or table stakes that customers assume. But when the lights fail, the room isn’t clean and the coffee is cold, customers are immediately displeased.
Most approaches to measuring key drivers don’t differentiate between the upside and downside; they treat good/bad simply as opposite ends of a continuum. In their world, if the temperature of the coffee is a driver, cold is bad and super-hot is good. But it just isn’t so. Delighting and upsetting customers is not a simple function of performing better or worse on the same criteria. More often than not, the criteria are different. As such, the upside opportunity drivers of loyalty/satisfaction/promoters must be measured and managed separately from the downside risk drivers of dissatisfaction.
9. People Are Minimizers, Not Maximizers
We instinctively prefer to minimize our losses rather than maximize our gains. People react far more strongly to losing something they have than to gaining something comparable. A $5 increase in the copay on health insurance (that is, a loss of a $5 value), for example, elicits far more adverse reactions than a $5 decrease in copay (a gain of a $5 value) generates positive reactions. JCPenney might have considered this before they abandoned their strategy of never-ending sales and switched to EDLP (everyday low pricing). For its longtime customers, this meant the loss of sales prices and savings more than the gain of lower prices overall.
10. The World Is Non-Linear . . .
Most approaches to measuring and managing customer loyalty and experience are linear. Everyone says the world is not linear, after which most retreat to regression-based explanations using “line of best fit” solutions. I’m pretty certain that lines are linear. The rational decision-maker of traditional economics is linear. In that world, businesses have a simple challenge: develop proof points to demonstrate the superiority of their products and services;disseminate; done. There is no real need for marketing, just facts and proof.
Reality is far messier. In the domain of behavioral economics, the consumer is more complex. In this realm, consumers are partially or quasi-rational with limited access to/ interest in/understanding of information. The behavioral domain presents a more difficult challenge of maintaining, reinforcing or breaking behavior patterns in minds.
11. Everything Is Relative
The only absolute is that there are no absolutes. People need context. That context is the ruler against which they rate performance and make choices.
Meaningful measurement, as such, requires a meaningful ruler and context. A ruler of unspecified length and without markings provides little guidance. Elongated rating scales without anchored points are invitations for guesses in a vacuum. Such scales increase variance (good for regression modeling) by virtue of increasing inconsistency across respondents and even across questions for the same respondent (bad for accurately understanding what drives behavior). This scale usage heterogeneity, as the stats people call it, can dramatically distort the data.
Whenever possible, my preference is to ask people to rate performance against their expectations using fully labeled scales. While expectations will vary between people, this provides a frame of reference that is most meaningful for the individual respondent. Our opinions, assessments and decisions are inherently comparative— and VoC measurements and programs should reflect this reality.
Where Do We Go From Here?
Two paths diverge in yonder woods… one returns to the traditional economic models of rational decision-making in a linear world, and the other looks to adopt measurement and management practices to the far more complex world of behavioral economics. While dealing with change in the face of corporate politics, inertia, legacy and the inherent tendency for CYA can be challenging, choose your path wisely.
You can read an expanded version of this piece at Loyalty Leaders.