In the early 1980s, the music division of Warner Bros. commissioned a special marketing survey to understand a paradoxical trend in its consumer behavior. Market surveys revealed that older people who were strong buyers ardently disliked the music they repeatedly purchased. This finding flew in the face of the traditional belief in the sales and marketing field, i.e., the idea that people’s behaviors can be predicted from their overall attitudes. Probably due to its immediate face-validity, the belief in a direct link between attitude and behavior persists to this day.
Yet, since the mid-1970s behavioral psychologists have known this link to be grossly incomplete, if not downright incorrect. And, when their insights were applied to Warner Bros’ puzzling consumer behavior, researchers learned that parents and grandparents were buying music they personally disliked as gifts for their children or grandchildren. In other words, consumers were not simply motivated by their own attitudes, but in this case their actions were determined by the wishes of loved ones.
This finding made perfect sense from a motivational point of view. Behaviors are determined by intentions, in turn these intentions depend on people’s attitudes and social norms regarding a particular behavior. Arguably, the Warner Bros. market research was the basis for one of the first “targeted ad” campaigns based on progressive yet rigorous science. The classic slogan born from that research is the instantly recognizable call to action – “Give the gift of music”.
Moving on From Attitudes to Psychological Drivers
Surprisingly, few marketers learned the critical lesson from the Warner Bros. case study. That lesson is that attitudes are expressions of deeper, psychological drives. And those psychological drivers are what businesses need to measure and model if they wish to impact consumers.
Rather than identifying key drivers, consumer research traditionally stops when it identifies an interesting cluster of behavioral patterns. The static PRIZM market segmentation analysis is perhaps the most popular and widely used example of this approach. Nowadays, static models have become the basis for staple, real-time “recommendation engines” used by companies like Blockbuster, Amazon.com and eBay. Such companies bank on return business, under the assumption that past behavior is the best predictor of future behavior. This thinking also dominates social networking and e-commerce websites, as well as online dating sites that profit from online advertising revenue. A good example can be found in the Associated Press story (2) about the targeted ad program recently launched by social networking giant Facebook which uses past behavior as its main variable.
Although approaches like the Facebook program show that old habits of businesses are hard to break, the same cannot be said for customers’ behavior. That seems like a surprising claim to make given that consumer behavior often appears habitual. For instance, despite some well designed campaigns, most current smokers continue to smoke. Are such behaviors “habits” – and thus seemingly removed from rational consideration – or are they in fact under rational control? Rense Lange, Ph.D., a member of the 20|20 Skills™ assessment team and one of the market researchers on the Warner Bros. project, points out that years of research in social psychology and marketing shows that our behavior is to a large extent determined by rational factors, including our beliefs, evaluations and intentions, as well as social pressures. As an example, most smokers will attest that in their case continuing to smoke seems to have fewer negative short-term consequences than does quitting. Thus, many smokers continue to smoke because to them it is a rational choice.
Does Buying History Really Repeat Itself?
In practice, those popular “recommendation engines” work something like this:
- Consumer A made a specific purchase or showed particular consumer trend [observation]
- Consumer A therefore probably likes these types of products or services [once-removed assumption]
- Consumer A therefore will probably want these types of products or services in the future [twice-removed assumption]
- Consumer A therefore will probably spend money on similar products and services [third-removed assumption]
Is this reasoning really the best predictor of future consumer behavior? The answer is no. People are creatures of habit in many respects, but this does not imply that they are unaware of such behavior or that their repetition is a reliable indicator. Repetition (e.g. in the form brand-loyalty) simply reflects that people did not receive sufficiently compelling information that causes a consumer to revise their prior behavioral beliefs.
By contrast, the “consumer behavior as habit” notion assumes that routines and rituals provide a sense of familiarity, comfort and control. In essence, consumer “habits” are seen as providing a form of equilibrium. But this equilibrium may be illusory. Physicist Per Bak formulated the notion of self-organized criticality in physical and physiological applications (1), and the literature increasingly shows important applications of criticality approaches in economics and psychology (5). Self-organized criticality views nature as perpetually out of balance, although temporarily organized in a stable state where change is possible according to well-defined statistical laws. In other words, the basic theory that that equilibrium or balance is the norm throughout nature is rejected. Instead, Bak proposed that systems with many components are actually in a state of constant imbalance. Hence, nature (including consumer behavior) is not static, but rather a dynamic, ever-changing process. The same appears true for people’s beliefs and attitudes, but fortunately advanced mathematics can model the process and identify ways to influence it (3).
This leads to an important point, namely, the “consumer behavior as habit” notion is inherently fatalistic. Intervention is impossible, since the habit notion implies that old habits are static, overly rigid and difficult to break. Yet, we have found that behaviors can be changed by giving tailored information that succeeds in changing consumer’s relevant beliefs, attitudes and intentions over time. Consumer behavior changes concomitant with these variables, and thus becomes dynamic rather than static.
Modern Analytics Meet Progressive Consumer Models
In contrast to the above, some businesses that are willing to abandon simplistic notions have invested in progressive research to “read the minds of consumers” in ways unimagined even a decade ago. Designing the right marketing methods and messages may not be easy, and it may require considerable research, but it can almost always be done when these insights are kept in the forefront:
- State-of-the-art mathematical approaches like Item Response Theory are crucial for identifying and measuring accurately consumers’ hidden preferences and motivations.
- Specific consumer behaviors are highly predictable when the hidden motivations behind those behaviors are understood.
- Consumers’ hidden motivations encompass far more than simplistic attitudes or habits.
- Carefully customized messaging is needed that takes qualitative and quantitative consumer differences into account.
- A variety of approaches should be used (e.g., particular “calls to action,” use of specific colors, identification of counterintuitive behavioral predictor) to reinforce these hidden consumer motivations, and thus provide the most qualified leads.
Take a Step Back Before Moving Forward
Market studies, especially those geared towards targeted advertising applications, should be done well or not done at all. It is a wasteful to conduct market studies of any type based on outdated and misguided consumer models and using imprecise statistical approaches.
On the other hand, the ideas and approaches outlined here offer tremendous potential for reaching many business goals, such as:
- Identifying the hidden motivations behind buyers and non-buyers of your products or services.
- Developing a sales-marketing campaign to attract and influence the consumer behavior of a specific demographic.
- Gaining empirical insights into strategies for improving your brand awareness, equity and quality, especially within certain demographic groups.
Understand upfront that studies that aim to “read consumers’ minds” at this level are resource-intensive. Yet, forward-thinking organizations can obtain a powerful competitive advantage with the right investment in this line of research. Warner Bros. reaped the rewards of what was essentially a first-mover advantage over twenty years ago, but statistical modeling never stopped improving. Applying modern analytics to progressive consumer models is the next generation strategy. And to think you may have been under the impression that “recommendation engines” were the best contemporary research had to offer!