The Second Truth in Advertising

THE SECOND TRUTH IN ADVERTISING
Media Chads and Other Recounts

Those of us in advertising (or politics) are concerned with truth - both kinds. There is the side of truth that the consumer sees, reads or hears in the advertisement itself. This is the public truth (or sometimes exaggeration) that has been much written about and self-regulated. It is what we commonly call "truth in advertising." Then there is the other truth.

We practioners grapple with the other truth every day. That truth relates to the underlying principles we employ to read the marketplace, construct strategies and deploy advertising. It is the truth that advertising research has tried to reflect from the beginning. The problem is, no matter how truthful we try to be, the reflection of this truth is often seriously distorted. The truth is… we only have an idea of how serious the flaws are. What we have here is one giant media chad and it is not confined to a few counties or one state or even half the population that registered to vote. It relates to the entire US population, over the age of two, that buys, reads, watches, listens or breathes and is surveyed. In other words the surveys we depend on are seriously flawed.

Chaos in Media

In the early 1960s, Edward Lorenz, a mathematician at MIT, had begun to codify the progress of weather patterns on his computer. He reasoned that if you could capture these patterns in equations, you could develop a fool-proof method of measuring and predicting the weather. The problem was, he rounded his equations off to the nearest thousandth (which is what weather satellites did). It turns out that he really would have to not round off at all, because even a slight variation beyond that rounding would create vastly different results as more and more time passed. This was a little like starting two overlaying straight lines at the same point and heading them off in the same direction, only one of the lines aims infinitesimally more to one way or the other. The further you move away from the starting point, the more the lines diverge. So if a current condition wasn't codified numerically to an infinite number of places, he could not predict the weather a month or even a week from now.

This diverging phenomenon became known as the butterfly effect. A butterfly flaps its wings in China and if it is not recorded, eventually that puff of air and all others like and unlike it would create vastly different weather patterns as time passed, like diverging lines. This discovery led to the science of Chaos, which is both a branch of mathematics and seemingly the constant state we live in today, professionally and personally. The surveys we use to measure all manner of buying goods and consuming media are subject to the whims of Chaos. Moreover, consumer behavior could not be plotted on a straight line. There is nothing linear about how people buy, watch, read and listen.

Sampling Error

If that weren't enough, sampling error plays a role in the fragility of surveys as well. For example, if one wanted almost certainty in their sample results, a cost of $75 to buy a rating point of a fringe television schedule in a medium sized market against adults in is really between $15 and $135 (a swing of +/-$60). Yet buyers negotiate for and are held accountable for…pennies.

Reconsider, Don't Recount

No matter how many times we count these things, the problem doesn't go away, so now what? For one thing we can stop pretending that we can measure public behavior so precisely. No matter how many chads (intentions) and recounts (sampling error) and fine-tuning (sample weighting) are done with research, we need to recognize that this is a complex country. If our system for electing a President is flawed on an essentially simple A/B choice, imagine how many chads would be lying on the floor and how many recounts would be necessary to get consumer behavior right. At some point, every practitioner must "trust" their own experience and that of others who have been hired to interpret information and make decisions and to know what they are doing. After all, expert experience is born out of thousands of surveys taken over time and the results that eventuated. Reading the market, is still the province of experts and we need to reconsider the standards to which we hold them accountable, while we respect the role of research.