Dorothy Leaves Kansas: A New Approach to Marketing in the Membership Resort Industry

This article’s title is suitable in that what is covered in the following paragraphs—Impulse Propensity Marketing—is as much a departure from the industry’s more prosaic marketing practices as was Dorothy’s trip to Oz. Baum’s narrative was fictional, however, while this one, rest assured, is not.

Impulse Propensity Marketing or “IPM” is a field-tested, permission-based, demand-generation methodology premised on the theory that pleasure-related consumer behavior begins primarily on an emotional level…and is rationalized secondarily. As consumers, people are feeling creatures that happen to think.

The point of the spear for this research-supported theory is a predictive model incorporating a framework of risk-propensity segmentation driven by attitude-and-behavior psychographics which together provide the marketer with the potential to behaviorally classify and commercially manage less-cognitive, pleasure-seeking impulses. This effort to “catch lightning in a bottle,” a phrase attributed by many to the baseball great, Leo Durocher, is not at all as quixotic as it may at first seem…but its contemplation does require a degree of open-mindedness.

In an IPM campaign, consumer participation is reactive and incremental. It begins as a low-involvement attraction to the marketer’s premium or giveaway inducement—a car, for example, or money, or a trip—as displayed in a mall, box route location, or internet advertisement, where the marketer’s only goal is to offer the prospect an appealing invitation to “opt in,” sign up, or otherwise register with the permission-based promotion. At this stage, a minimum amount of information is obtained: key demographics, the permission block, and the prospect’s responses to 4-6 disarmingly innocuous impulse propensity questions. It should be pointed out that the “hard” information collected is of a minimally invasive nature. Once the “lead” is collected, impulse propensity classification and prospect screening take place within data management, the call center, and confirmation, thereby eliminating the telltale push-me, pull-you of traditional prospect registration that has for years fueled the decline in giveaway participation by credit-worthy consumers.

Operationally, an impulse propensity classification is appended to each record in the prospect data file and follows the prospect throughout the entire handling chain from the phone solicitation, to the tour, to the sale, to service, to receivables…and even to the re-marketing campaigns for trips back to the well with the existing member base. At each point of contact it is possible to know the impulse propensity, hence the behavioral class probability, of the prospect, and to adjust the marketing and sales process accordingly. Note, too, that it also is possible to use this classification in conjunction with wealth class segmentation systems of geo-demographic profiling for highly targeted direct marketing.

When it is the goal of the marketer to improve the creditworthiness of the prospect pool, the ability to predict behavioral class becomes essential. Why? Because most lead allocation schemes are premised on random distribution, which is to say that whatever proportion of the prospect pool is made up of wealthier (and more sales-resistant!) consumers, most will be assigned by random distribution to that larger segment of the Call Center or Sales staff that in reality is less-prepared to effectively convert prospect to purchaser. The most important stage in the entire prospect handling chain traditionally has been left simply to chance! However, because IPM enables Call Center lead assignment and Sales Center prospect assignment to that portion of the staff with greater conversion skills, it offers the potential for an improved bottom line with overall lower costs in demand generation: better leads, smaller staff, more sales, higher sales volume, and greater profit.

Related: Premium Perception Disconnect Part 2: Wrestling with Squirrels

What proof is there that IPM can work? Proof obviously is a problematic word, but let’s take a swing at it.

Confirmatory Factor Analysis (CFA) is used to generate a model of the data, portraying a set of five attitude-and-behavior variables drawn from the full inventory of twenty that were employed in the pilot study. (A copy of the model is available, gratis, by contacting the author.) Selected on the basis of their significance in predicting a consumer’s capacity for impulsive behavior, these five variables may be grouped into two inversely correlated factors (r=-.81). Conceptually, one factor may be thought to house consumers more likely to behave in an impulsive way, while the other factor may be thought to house consumers who are less likely to behave so. The goodness of fit results for the model lie within widely accepted standards and are shown in the bar beneath the display.

Following the development of the 2-factor, 5-variable CFA model, cluster analysis (convergent K-Means, Ward’s method) was used to segment the data so as to create two respondent clusters comprised essentially of a set of respondents appearing to be more impulsively inclined, and a set appears to be less so.

In the pilot study of more than 700 respondents, approximately 370 respondents fall into the “more impulsive” class, and a similar number fall into the “less impulsive” category…so there is an inherent or “state of nature” probability of about 50% that a randomly selected respondent will come from either class. Literally, when nothing else is known, the random selection, or guess, in such a scenario will be correct about half the time. However, when the 5 variables are incorporated into a logistic regression, the “hit rate,” or probability of identifying the correct behavioral class, improves by nearly 30 points to 80%! While more work lies ahead, results like these are extremely encouraging.

One of the 20th century’s preeminent statisticians, George Box, suggested that “All models are wrong, but some are useful,” and that sage insight serves well as a cautionary against blind reliance on statistics and all of the arcane mumbo-jumbo that occasionally leads marketers down the wrong path (think, New Coke). But it is nonetheless clear that while the ultimate provability of any given model may be subject to debate, the argument in support of consumer marketing premised on the theory of impulse propensity is clear, cogent, and compelling.

D & A Solutions, Ltd. is a boutique consumer research firm specializing in statistical analysis of Voice of the Customer surveys and enter-to-win Permission Marketing Lead Generation for the timeshare and membership resort industry. You may reach the author, Ken Will, Director of Consumer Research for D & A Solutions, Ltd. At (330) 526-6954, or email him at kwill138@neo.rr.com. The D & A blog: http://dandasolutionsltd.blogspot.com/

Originally printed in March/April Resort Trades Management & Operations