The affinity diagram is a recognized “best practice” in
New Product Development, earning a reputation as an
efficient and intuitive tool for organizing customer needs
into a logical hierarchy for product planning purposes.
In their landmark article, “The Voice of the Customer”
(1993), Professors Abbie Griffin and John Hauser
concluded that involving customers in creating the
affinity diagram was unequivocally better than leaving it
to the internal product development team – a common
practice both then and now. Taking their advice, many
companies began using customer focus groups to build
the needs hierarchy. Yet Griffin and Hauser suggested
a more rigorous process where a large sample of
customers would individually create affinity diagrams,
which would then be statistically merged using cluster
analysis. This method would reflect the independent
thinking of a larger number of individual customers
and would avoid the “groupthink” mentality inherent
in focus groups. Moreover, by asking participating
customers to choose an “exemplar” need statement
from each of their individual groupings, the process
would provide concrete information from which to
name each of the resulting clusters. At AMS, this
process became VOCALYST®.
Following the explosion of online market research, AMS
began to explore moving this exercise to the web. Our
goal, first and foremost, was to retain the integrity of
Griffin and Hauser’s methodology; online affinitization
needed to yield the same level of precision as the
original model using paper cards. After years of
research and testing various prototypes internally and
with several clients, we began offering online
affinitization to our VOCALYST clients, introducing
several important benefits, including:
- Faster completion times. The use of paper cards typically
requires telephone recruiting, followed by mailing a packet of
cards, reward money, and a postpaid envelope for the respondent
to return their sorted cards to AMS. The data are then entered
manually. Turn-around time for the whole data collection process
often exceeds two weeks. On the other hand, web-based cardsorting
lets respondents complete the exercise in about a day or
two, and because paper-shuffling is eliminated, we retrieve the data
instantaneously in electronic form and can begin analysis quickly.
|
|
- Easier recruiting. Thanks to the growing multitude of opt-in
web-based panels (i.e. people who agree to participate in
market research studies in exchange for cash or prizes), we can
now recruit for many studies using e-mail. Whereas telephone
recruitment typically takes one to two weeks and requires the
services (and expense) of a live recruiter, e-mail is automated,
letting us complete recruitment in a matter of days or in some
cases, hours.
- Lower costs. Online card-sorting eliminates the administrative
expenses of printing, mailing, postage, and data-entry, and yields
significant savings in respondent incentives (the reward paid to
those who participate). Web-based card sorting allows incentives
to be paid only to those who complete the exercise, rather than
mailed to all who agree to participate at the outset. Furthermore,
because participants are already enrolled in a panel, the incentive
required per respondent can often be less.
Despite its clear benefits, online card sorting is not
appropriate in all situations, such as when much of the
target population is not web-literate – for instance, the
elderly, the less educated, or people in developing nations.
Also difficult are narrowly defined populations with highlyspecialized
skills or training, as is the case with certain
medical specialties. Nevertheless, this method is proving to
be a major advance in our VOCALYST® service – a classic
case of “better, faster, cheaper” results for our clients.
—Gerry Katz gkatz@ams-inc.com
© Copyright Applied Marketing Science, Inc.
|