Yet return on investment has become the driving obsession of the marketing industry, igniting as much passion as anything involving econometric modeling ever could.
A year after a report by Deutsche Bank concluded that advertising doesn't generate positive ROI for most mature package-goods brands, researchers are still debating its merits and trotting out competing analyses. Tens of millions of dollars and thousands of Portable People Meters are being deployed in the name of improving ROI. CEOs can rarely escape an investor conference call without fielding questions on the subject or offering peremptory defenses of their marketing ROI performance.
A basic problem is ROI remains poorly defined, says Erwin Ephron, principal of Ephron, Papazian & Ephron. "It's come to mean any measure of accountability."
And increasing demands for accountability aren't necessarily translating into greater understanding of marketing results. A survey of Association of National Advertisers members conducted by Forrester Research and Aegis Group's Marketing Management Analytics found only two in five marketers could project the impact of a 10% cut in marketing spending.
"When the CEO asks what happens if I cut 10% and you say you don't know, the next step is he wants 20%," says John Nardone, exec VP-chief client officer at MMA.
POOR RESPONDERS TO ADVERTISING
Another big problem for ROI is that the marketers who understand it best are the ones that get the least of it. The package-goods industry, focus of the Deutsche Bank report, is blessed with the most elaborate array of analytical tools but some of the least advertising-responsive brands and businesses to be found in marketing, Mr. Ephron says.
Package-goods brands tend to be mature and smaller than $1 billion, he says, making the ROI job exceedingly hard for their marketers and agencies. Billion-dollar brands, even mature package-goods ones, almost always have positive ROI from advertising, regardless of how good or well-placed it may be, he says.
Unlicensed drivers of marketing-mix models are wreaking havoc, Mr. Ephron believes. As big marketers like Procter & Gamble Co. drive use of econometric models down through the ranks, they're being used and interpreted by junior marketers only dimly aware of how they work.
Even given a common definition of positive ROI-at least a dollar of net profit generated per dollar spent on marketing-the numbers don't, or shouldn't, necessarily mean what many marketers or financial analysts think, Mr. Ephron says. "You may see that the ROI of TV is 59¬Ę and the ROI to print, which you've used sparingly, is $1.23," he says. "But you don't know what the shape of the print curve is. If you double spending, the ROI may go down to 59¬Ę."
Another common fallacy, Mr. Ephron believes, is seeing a high ROI as good, when what it really means is that the marketer hasn't spent enough on a medium. "An ROI of $1.50 to a medium means you have not really spent enough money in the medium, so it's not a good solution," he says. "What you really want in a rigid ROI model for everything to come out at 1 [$1 return per $1 spent] because the real goal is to grow the brand, and if you can do that at no cost, that's the best position because you're not in the business of making money on media or on other marketing inputs."
Marketing-mix models-and in particular the Deutsche Bank report-don't fully take into account long-term effects of advertising or the halo effect that advertising one product has on other products in a brand lineup, said speakers from Coca-Cola Co. and PepsiCo's Frito-Lay at an Advertising Research Foundation presentation in April.
About half the brands in the Deustche Bank research showed positive long-term ROI from advertising, based on various studies showing the long-term sales lift is double what occurs in the first year after ads run. But the real long-term effect may be as high as three times first-year results, Ravi Parmeswar, VP of Coca-Cola USA, and E. Craig Stacey, director of the Zyman Institute at Emory University, said at the presentation.
As far as halo effects, the positive impact that an ad for, say, Tostitos Scoops has on the entire Tostitos franchise can bump long-term impact of advertising as high as four times the first-year impact on sales, says Dwight Riskey, senior VP-consumer insights of Frito-Lay.
But Andrew Shore, the analyst who headed the Deutsche Bank study and now works on the investment banking side of the house, says his ROI research did in fact look at long-term and halo effects, and still found advertising lacked payback for most mature package-goods brands.
"If it doesn't work in the short term, it doesn't work in the long term," he says. "We're not saying advertising absolutely doesn't move volume. The real point is that [marketers are] spending way too much, and it's time that they get their arms around it."
P&G has put a major emphasis on finding ways to quantify and improve its marketing ROI in recent years. Tapping Joe Auriemma, whom P&G brought in with its 2001 acquisition of Clairol, the marketer has expanded use of marketing-mix models to nearly all brands. Mr. Auriemma now holds the P&G job title of director-consumer and market knowledge.
In addition, P&G has applied a technique, originally tried in Argentina, using actuarial loss analysis, essentially treating marketing waste like an avoidable accident. And it has charged Don Gloeckler, manager-North American media research, with improving media metrics in the U.S., pushing-so far unsuccessfully-for affordable and reliable minute-by-minute TV commercial ratings, he told the ARF conference.
"Every brand in every country that has done marketing-mix modeling or marketing ROI [analysis] has improved the productivity of its marketing spending without exception," P&G Chairman-CEO A.G. Lafley told analysts in an April conference call. "We're training [marketers on] it like crazy. We're finding appropriate, what we call `light' versions to get into developing markets. ... I think we're going to benefit from this for at least another two or three years."
For those who foresee trouble for media budgets from increased use of marketing-mix models, it's worth noting that despite all the analysis, P&G's ad spending has risen for the past three years. Then again, P&G also has cut its upfront commitments this year.
Court TV is conducting an ROI study with media shops. Focus areas include: Will ROI efforts change how platforms are bought? How will ROI measurement efforts change spending both on network and cable TV?
ROI often MIA
While the term "ROI" is popping up with disconcerting frequency among marketing executives, a survey of Association of National Advertisers members reveals that application of ROI principles is lagging. Among the study‚Äôs findings:
* Only 1 in 3 marketers built marketing budgets based on knowledge of spending required to meet corporate goals.
* Only 13% of senior marketers felt confident in their ability to forecast sales impact of their marketing programs.
* Faced with a 10% budget cut, only 2 in 5 felt they could predict the sales impact.
Source: Marketing Management Analytics