I suspect that more time is given to debates over whether
and how to segment an audience than to any other decision made in a social
marketing program. And for good
reason. The process of segmentation
distills the aspirations and predilections of stakeholders and designers into
an essence that will (or a least should) permeate every aspect of the program.
Audience segmentation operates from the principle that
"birds of a feather flock together." Each of these flocks share certain characteristics whether they be
sociodemographic or psychographic (or choose your own). However, each flock is distinct from the
others based on other variables. What we
are doing in social marketing is trying to figure which flocks are most
important to target to improve public health and social welfare, usually around
a specific topic (breast cancer prevention, recycling, participating in
mentoring programs). In the marketing
business, these flocks often go by the name of “target audiences.” While the notion of a “target” makes sense in
the militaristic jargon that permeates marketing, I often find that the term
“priority audience” gives a sense of reassurance to practitioners and
stakeholders that we are not somehow dehumanizing the people we serve and, on
the other hand, recognizes that - as in all things - priorities (including
audiences) can shift over time.
Many writers in the social marketing world (including
myself) have advocated for the creative use of many different types of
segmentation strategies to design programs with more relevance, greater reach
and increased effectiveness. Indeed, the
mainstream thinking among commercial marketers today is to aim for the
“audience of one” through what is known as mass customization. Whether or not this segmentation strategy
will prove practical and effective for social marketers remains to be
seen. However, what is often a reality
for most social marketing programs is that limited data about priority
audiences, the lack of expertise in using multivariate statistical analyses to
develop segmentation categories, resource restrictions that hamper the
tailoring of the marketing mix to multiple audiences, and mandates to achieve
impact across a broad swath of the population make the use of sophisticated and
fine-tuned segmentation strategies difficult to implement.
In some of the recent work I have been doing, we have been
looking at the question of audience segmentation from the perspective of
addressing three basic questions:
Who are the people at
highest risk? This question is
designed to mine the demographic and epidemiological data that are often the
only ones available to program designers. The answers to this question can then be combined with other data
collected during the formative research process to identify one or more groups
of possible priority audiences.
Who are the people
most open to change? From among the
groups identified in response to the first question it makes sense from both a
theoretical perspective (e.g., diffusion of innovations, stages of change) and
a practical one to focus our initial efforts on those subgroups that are more
predisposed or motivated to engage in new behaviors.
Who are the
people/groups critical to the success of the program? Social marketers usually rely on
other people or groups to implement various parts of the program (peer
influentials, intermediary organizations, media representatives). At other times, policy makers or senior managers
may be key determinants for the long term sustainability of a social marketing
effort. Yet, rarely do I see program
designers explicitly focus a marketing strategy on these types of audiences
despite their often critical role in accomplishing the goals of the program.
Depending on the marketing (behavioral or social change) objectives of the program, there may be more than one answer to each of these three questions. What we often find in going through this exercise is that question #2 (who is most open to change) usually fine tunes the choices made to question #1. While this 2-step process is the more obvious part of segmentation, I will submit that it is in the ignoring of the last question where most programs discover (too late) their plan’s Achilles' heel.