|
6. The Caples Testing Method
The United States in the l92Os was bursting with eager,
upwardly mobile people, many of them first - and second- generation immigrants
determined to make good in the New World. Books, courses, and other methods
of self-improvement were very popular and were widely sold by mail-order
advertising, an ideal method of communication and distribution for that
kind of product.
To accommodate these important advertisers, an increasing
number of newspapers and magazines offered split run service. The advertiser
could supply two different advertisements, with the reply address in each
keyed differently, and the publication would run both ads in the same
issue and position.
In a true A-B split, every other copy of the publication
would contain ad A, and every other copy would contain ad B. This advertising
research technique met the same requirements of the scientific method
that Luther Burbank used in comparing the germination of two seeds under
identical conditions.
Caples was apparently so eager to get on with talking about
and showing what works in advertising that he did not vigorously crusade
in his book for applying this research technique to awareness advertising
that does not pull for an immediate order.
But he did provide examples of it, "Perhaps in years to
come, more advertisers will use scientific methods," mused Caples. "Perhaps
more advertisers will run tested copy in tested media. In answer to the
question, 'What will advertising be like 30 years hence?' a famous advertising
man replied, 'It will be more exact, more scientific, and therefore more
resultful."
How the Caples Method Works
The essence of the Caples method of scientifically comparing
two advertisements is to vary the advertisements - a different promise,
a different price, a different method of presentation, even a different
product name-but make an offer which remains identical in wording and
presentation in both advertisements.
The offer-which could be a booklet, sample, savings coupon,
or simply an offer to supply the nearest dealer's name-can either be featured
in a reply coupon or buried in the text. (Or it can invite a phone response
via an 800 number.)
A coupon will produce more replies than a buried offer and
thus lends itself to more statistically significant measurement. However,
the buried offer ad will more nearly simulate what the same ad would look
like without an offer.
The Caples method remains a valid way to test different
appeals, product attitudes, and methods of presentation in awareness advertising
today.
The testing can be done in any medium where absolutely equal
A-B splitting is possible. Such testing lends itself ideally to direct
mail, where it is a simple matter to create any number of test cells with
identical consumer characteristics. A-B tests of magazine, newspaper,
and television advertising can be devised almost as easily.
Adherents of traditional methods of marketing research might
argue that the 1 person out of 1000 who responds to a buried offer differs
from-and does not represent the whole. And this may have some truth.
But the same could be said of the consumers in a small-city marketing
research like Pittsfield, Massachusetts, who are observed by traditional
research methods. Consider also that many of the other 999 out of 1000,
whose opinions and reactions are studied and observed by traditional marketing
research, are not serious prospects and will never become serious prospects.
So the reaction of the targeted individual who is the reason
for the advertising in the first place should be more significant than
the reaction of a non-prospect. And the Caples method has the great virtue
of being a REALITY-BASED reaction in a real advertising environment rather
than an artificial focus group laboratory experiment or observation in
a theater or shopping mall.
It is fascinating to observe in Caples' process an example
of a split-run test of two different appeals for a product that is still
alive and kicking today, Milk Bones, the snack food for dogs. Both ads
carried coupons offering a free gift package of Milk Bones Snacks. Both
ads carried the same message that feeding your dog table scraps in the
summer was dangerous, and that you should give your dog healthful Milk
Bones instead.
One ad used a negative appeal and the other a positive appeal:
Ad A: Don't Poison Your Dog!
Ad B: Keep Your Dog Safe This Summer!
When the coupons were counted, they found that ad B had
pulled 58 percent more requests.
Do you find it hard to believe that this was a meaningful
result? If you were the marketing director for Milk Bones at the time,
wouldn't you have found this to be extremely useful information? Wouldn't
it have been helpful in developing a sound copy platform?
And wouldn't the information cost far less to develop today
than dubious mall interviews or eye-movement measurements? Doesn't it
defy common sense to believe that the dog owners who did respond to the
offer reacted somehow differently to the two appeals than the dog owners
who did not?
|
