Selecciones - Webs

Belch: Advertising and
Promotion, Sixth Edition
VI. Monitoring, Evaluation,
and Control
19. Measuring the
Effectiveness of the
Promotional Program
consideration. Ogilvy Award winner GM used focus groups to derive the concepts to
promote its new minivan. In other instances, layouts of the ad campaign that include
headlines, some body copy, and rough illustrations are used. For TV commercials,
storyboards and animatics may be tested. The GM minivan research also involved the
evaluation of six animatics. In these tests specific shortcomings were identified, and
the ads were changed to enhance certain executional elements.
The methodologies employed to conduct pretests vary. In focus groups, participants
freely discuss the meanings they get from the ads, consider the relative advantages of
alternatives, and even suggest improvements or additional themes. In addition to or
instead of the focus groups, consumers are asked to evaluate the ad on a series of rating
scales. (Different agencies use different measures.) In-home interviews, mall intercept,
or laboratory methods may be used to gather the data.
The advantage of pretesting at this stage is that feedback is relatively inexpensive.
Any problems with the concept or the way it is to be delivered are identified before
large amounts of money are spent in development. Sometimes more than one version
of the ad is evaluated to determine which is most likely to be effective.
A study of 4,637 on-air commercials designed to build normative intelligence conducted
by MSW Group (formerly McCollum Spielman Worldwide) found that only 19
percent were considered outstanding or really good. Nearly twice as many (34 percent)
were failures. On the other hand, of those spots that were pretested before the
final form was aired, the share of good to outstanding rose to 37 percent, while the failure
rate fell to 9 percent. 12 This is certainly a testimonial to the value of pretesting.
The disadvantage is that mock-ups, storyboards, or animatics may not communicate
nearly as effectively as the final product. The mood-enhancing and/or emotional aspects
of the message are very difficult to communicate in this format. Another disadvantage is
time delays. Many marketers believe being first in the market offers them a distinct
advantage over competitors, so they forgo research to save time and ensure this position.
Posttesting Posttesting is also common among both advertisers and ad agencies
(with the exception of testing commercials for wearout). Posttesting is designed to (1)
determine if the campaign is accomplishing the objectives sought and (2) serve as input
into the next period’s situation analysis. An excellent example of using research to guide
future advertising strategies is reflected in an experiment conducted by Lowe’s, the
nation’s second-largest home improvement retailer. In a study designed to test 36 different
versions of covers for its catalogs (which are sent to between 30 and 40 million
homes per year), the company determined that by putting more products on the covers,
using real pictures rather than cartoons, and reducing the size of the catalog, the catalogs
were more effective. Other tests varying the number of TV spots, newspaper ads, and
sports sponsorships led to increases in advertising spending and affirmation
of the company’s sponsorship of NASCAR auto racing (Exhibit
19-3). 13 A variety of posttest measures are available, most of which
involve survey research methods.
Where to Test
In addition to when to test, decisions must be made as to where.
These tests may take place in either laboratory or field settings.
Laboratory Tests In laboratory tests, people are brought to a
particular location where they are shown ads and/or commercials.
The testers either ask questions about them or measure participants’
responses by other methods—for example, pupil dilation, eye tracking,
or galvanic skin response.
The major advantage of the lab setting is the control it affords the
researcher. Changes in copy, illustration, formats, colors, and the like
can be manipulated inexpensively and the differential impact of each
assessed. This makes it much easier for the researcher to isolate the
contribution of each factor.
The major disadvantage is the lack of realism. Perhaps the greatest
effect of this lack of realism is a testing bias. When people are
© The McGraw−Hill
Companies, 2003
Exhibit 19-3 Research
affirmed the value of a
NASCAR sponsorship for
Lowe’s
627
Chapter Nineteen Measuring the Effectiveness of the Promotional Program