Targeted Marketing and Predictive Models in Mature Market Research
Over the next 20 years most of the 74 million Baby Boomers will retire. This is almost twice the number of retirees as the previous generation, the Baby Boomer’s parents. Furthermore, TW+A Research estimates that the Baby Boomers are at least twice as likely as their parents to move to a new state after they retire. This means there will be twice as many Baby Boomers who will be twice as likely to relocate to a different state after they retire. Thus the active adult housing market for Baby Boomers will be at least 4 times larger then it was for their parents.
The size of this potential active adult housing market makes traditional marketing prohibitively expensive. Thus what is needed is a way to identify the Baby Boomers most likely to relocate to a community and development,
and to focus marketing efforts on these people. This is essentially the concept of targeted marketing. Crucial to targeted marketing is market research, the identification of those Baby Boomers most likely to move to a community and development (the target markets). Accurately targeting retiring homebuyers most likely to relocate can substantially increase the response rate to a marketing campaign and thereby significantly reduce the marketing cost per sale.
The determination of target markets for an active adult community is complicated because Baby Boomer retirees are amenity migrants, not sun migrants, like their parents. Silent Generation retirees (the Baby Boomer’s parents) who moved to a new state after retiring largely chose their destinations based primarily on weather, not activities. Baby Boomers, on the other hand, are choosing their retirement destinations based primarily on the activities they want to pursue during the “second half of their lives”. This explains the growing popularity of cold weather retirement destinations like Eagle River, Wisconsin, Hamilton, Montana, and Pagosa Springs, Colorado.
Two types of information need to be generated by the market research for a targeted marketing campaign to active adult homebuyers. First, because community amenities are what first attract new active adult home buyers to the area, it is necessary to determine a community’s amenity strengths that appeal to retirees in order to emphasize them in marketing materials. Secondly, it is also necessary to identify where the Baby Boomers most likely to relocate to the target community currently reside. This is of particular importance because unlike younger homebuyers, active adult homebuyers are not limited to where they can live by the location of their job.
Information on the Baby Boomer’s retirement decisions is available because about 18 million of them have already retired. In addition, the 8 million War Babies (those born between 1941 and 1945) who have retired are essentially indistinguishable from the Baby Boomers. In general, there are two distinct approaches to obtaining the information from these retirees that is needed to develop an effective targeted marketing campaign. One approach is to elicit information from selected people, generally by talking to them, conducting focus groups, or by administering surveys. The other approach is to use data showing decisions of War Baby and early Baby Boomer retirees in predictive models.
Both approaches have limitations. Surveys must include a representative sample and ask unbiased questions. Multiple choice answers often limit the responses and can be leading. On the other hand, open ended responses are difficult to quantify and interpretations of the responses are often subjective. The answers provided by those responding to questions must represent their true opinions, not guesses. Finally, because the representative sample of people from whom information is gathered in all likelihood cannot be asked the same questions at a later date, the results cannot be replicated. Thus answers to questions represent the inclinations of responders only at one particular point in time and cannot be used to predict their future decisions unless it is assumed that their situation does not change. Finally, since the person being asked a question has no investment in his or her response, the response may simply represent a whim of the moment.
Predictive models can identify retirees with a higher likelihood of moving to a community with particular amenity characteristics. However, they require data for their validation and implementation. They also require quantitative data, which may not always be available. Obtaining this data, especially the data needed for models predicting retiree behavior, can be time consuming and expensive. Finally, predictive models are based on decisions retiring Baby Boomers have made in light of the economic, demographic, geographic, and amenity characteristics at the time of the decision. By changing any or all of these characteristics, it is possible to predict changes in their behavior in different situations. Thus, predictive models reflect actual choices made by them not speculation on what they might do.
Economists use predictive models to explain consumer behavior, a product’s change in demand due to a change in price, for example, or to evaluate how a stimulus package will affect the economy. Predictive models are more accurate because they estimate people’s future actions on their past decisions, not what they say they might do in the future. Accordingly, TW+A Research uses predictive models to generate the information required for targeted marketing campaigns to active adult homebuyers. Furthermore, it is one of the few, if not the only firm producing mature market research to do so. – Gene Warren