“The key to success (maximum sales per dollar) lies in perpetual testing of all the variables”
The above quote by David Ogilvy, Chairman and Founder of Ogilvy & Mather (super successful advertising agency) lies at the heart of what is known as “scientific advertising”.
Invented by ad men of the early-mid 20th Century, and coined as a phrase by ad man extraordinaire, John Caples in his book of the same name, “Scientific Advertising” stakes its worth on refusing to make assumptions about what ads work and instead basing advertising decisions on the results of measurable testing done on previous advertising campaigns.
Back in the last century, when these pioneers were working, the Internet was not a factor, so they were mostly concerned with magazine, newspaper, billboard, radio and television advertising.
What they would do is at its heart quite simple:
Firstly, they forsook “artistic concerns” in favour of “results”. They wanted advertising that clearly and demonstrably resulted in increased sales, or more initial enquiries, or more responses to a free offer, or a healthy and profitable product launch. Fluffy concepts such as “branding” and “style” were not interesting or useful to these hard nosed ad men, They cared about “sales” and “return on investment”. This is what not only kept them in business but indeed made the best of them quite wealthy indeed.
What did they mean by “Scientific” Advertising?
Quite simply they used the word scientific to imply that their methods were rigorous, measurable and methodical. They applied a testing model to each advertising campaign in a similar way that a scientist would test any hypothesis to measure its validity. They were after empirical results and the unit they measured in was the almighty dollar.
What Does ‘Split Testing’ Mean?
Split testing, as stated by Ogilvy above, is the core methodology at the heart of scientific advertising … but what is it exactly?
It simply means that instead of running one ad, or one version of an ad, in one place (magazine, radio station etc) you split up your campaigns in various ways, in such a way that the results of the different campaigns can be gathered, measured and analyzed.
A very simple example of this is to run the same ad complete with a cut out coupon in three different magazines, and in the corner of each coupon including a serial number that identifies which of the three magazines the coupon was cut out of. Therefore, all things being even, if 65 coupons from Magazine A were returned, as compared to 370 from magazine B and only 28 from magazine C, then it is pretty clear which magazine is most worth running that advertisement in again.
Other variations on these basic themes exist, not least of which are headline variations. The reason headline variations are given a lot of prominence is because it is proven to be quite simply the single most important factor of any (written) advertisement.
