With ad spending expected to reach about $200 billion in the U.S. this year, this is a waste of about $100 billion. With at least $160 billion spent on media (and not on creative), about $80 billion of media spending is wasted. Yet we still don't know how advertising works, and we know much less about how media works.
BEST FOR NEW PRODUCTS
The first IRI study tells us mainly that advertising works best for new products and line extensions (where starting from nothing, any sale is a huge increase), and that small and medium size brands respond more than the big ones. It shows most of the impact of TV ads comes in the first four weeks of the campaign, and that there is a residual long-term effect and a slow decay once the campaign is over. It also finds that promotions can enhance TV ads and vice versa.
Preliminary results from this year's study find that half of the $14 billion spent on spot advertising is being wasted. It reveals that when most of the weight (and the money) is in prime time, TV advertising is most effective. It shows that share of voice is important-some of the time. And it also finds promotions can detract from TV advertising.
MARKETING MIX ANALYSIS
For the studies of "How Advertising Works," IRI and MMA use a marketing mix type of analysis. Originally employed to measure the relative impact of the main marketing forces on sales and to determine the return on the investment from each, it has been used effectively to reallocate marketing budgets. In the "How Advertising Works" studies, marketing mix analysis has been used to detect the relative impact of ad-related variables on sales.
Because of the data it collects (or adds, as with the data on specific commercials), IRI looked into advertising as a separated force from promotions, pricing and distribution. And within advertising, it looked into the use of different weights or gross ratings point levels (or spending), assorted dayparts, share of voice and spot vs. national spending.
But are these the only variables affecting the way media works?
INCREASING THE IMPACT
A large body of studies demonstrate that advertising in one medium in tandem with advertising in another medium can increase the impact of both. For example, various studies show magazine advertising enhances and adds value to the impact of TV advertising and vice versa. An understanding of the impact of one medium can thus be best understood in relation to advertising in the others.
And what about the way TV advertising works? Various qualitative measures have been the focus of attention in recent months, specifically to counter the optimizers-led impression that all rating points are the same. Some of these qualitative measures are "liking" and "appreciation" of TV programs, "attention" being paid to higher-rated TV programs and "involvement."
Some studies find the "liking" and "appreciation" of TV programs are similar to the programs' ratings, and increases and decreases in the "liking" and the "appreciation" of programs precede increases or decreases in these programs' ratings. Other studies show no such relationship. "Attention" to programs is found in some studies to be the highest among higher-rated programs. In others it is found to be the highest among low-rated, special-interest programs.
High "involvement" is found to increase the propensity of viewers to stay tuned during commercial breaks and make people more prone to be influenced by the message. Elsewhere, the relationship of "involvement" to commercial impact is found to have a U-shaped curve and that, in some types of programs, high "involvement" decreases, not increases, the ability of viewers to accept the commercial message.
(Still, other qualitative measures of TV programs are found to be highly discriminating and such variables enhance the impact of commercials embedded in these programs.)
MEASURING THE WASTE
While currently hailed as the way to identify the return on the advertising investment, marketing mix analyses can tell us how much of the money we waste. But it can't tell us how media works and how to do a better job of it in the future. It can't give us answers to questions we didn't ask, or tell us the impact of variables we didn't include.
To make media accountable, we have to be able to make predictions, to know what causes a particular outcome and how to effect or prevent its occurrence.
To tell us how media works, we need to include in the analysis all the media-related variables (including TV, but not only TV); employ additional analytical methods; better understand the relative impact of each of these variables; identify-or mathematically model-the relationship among them; and be set up and able to forecast differing future scenarios.
Because of the data IRI collects, its analyses do not include constructs related to other media, variables specific to the nature of TV or factors related to the many specifics of the media plans.
Marketing mix analysis may be the tool to tell us what was the return on our investment in advertising vs. that of promotions or other marketing spending. But, as is, it may not be the right tool to tell us how advertising and media work, or how to do a better job of it in the future.
THE EASY WAY OUT
As is, we are just taking the easy way out. It is like the story of the man who came out of a pub and saw another man on his hands and knees, crawling under the street lamp. "What are you looking for?" asks the first man. "For my car keys," answers the second. "And where did you drop them?" asks the first. "Over there," says the second, pointing down the dark alley. "So why are you looking over here?" asks the first man. "The light is better here," answers the second.
We still don't know how advertising works. We know even less about how "media" works. We know what the return on our advertising investment is, but we can't make media accountable unless we know more about how it works and how to do it better in the future.
For a possible waste of $80 billion in media, we should look into even the darkest alleys.
Ms. Bigman is president, Communication Options, New York, consulting on media research and development issues.