• by October 20, 2005

YOU CAN'T SPELL CONFUSION WITHOUT THIS -- What is it about fusion that always seems to cause such confusion? No, we're not talking about the promising form of nuclear energy, but the energy some people are promising to spend on a new form of marketing and media research. What's with all the fussing over fusion? Well, it's more about economics than it is about research. In an ideal world, marketers, agencies and media would spend what they needed to create the best research to make the most informed decisions possible. But this isn't an ideal world--at least not so far. In recent years, the trend has been toward pragmatism and compromise, not integrity and validity. The main culprit, of course, is cost. Or at least, that's the rationale used by organizations for cheaping out, and using methods and systems that, in the end, may yield less informed decisions, not better ones.

We can rattle off the litany if you care to hear it yet again. But we'll spare you that and zero in on two examples: ScanAmerica and SMART--and soon maybe even a third and fourth: the portable people meter and Project Apollo. Each was an attempt to develop superior methods of measuring consumer behavior that is the basis for billions of dollars of advertising and marketing decisions. The first two--Arbitron's pioneering single-source TV viewing and product usage measurement system, and Statistical Research Inc.'s innovative TV ratings laboratory--ultimately failed because of costs. Or at least, the perceptions of their costs. And there's a good chance the next two may be casualties of the same rationalization.

That perception of high cost is ironic, because marketers spend $60 billion on TV advertising buys in the U.S. alone. They spend more than a quarter trillion dollars on measured media advertising in the U.S., and more than a half trillion dollars worldwide. And that doesn't include all of the other marketing expenditures that are influenced either directly or indirectly by the building blocks of consumer audience measurement. So when people complain about better research costing too much, and when they opt to cheap out in favor of more economical approaches, we have to ask: What is the real cost?

We know that plenty of smarter research minds have said that before us and fallen on their own swords: SMART creator Gale Metzger and ScanAmerica champion Rick Aurichio, for two. And while he wasn't a casualty of his declarations, we always wondered why former MindShare share-minder David Marans got treated with a bemused sort of derision every time he got up in front of the Association of National Advertisers' annual TV Advertising Forum and made the case to spend a little more to get a lot more.

Marans' schtick was that if marketers would kick in and essentially double the amount of money spent on TV audience measurement, they would: 1) provide the resources necessary go get the kind of research they really want; and 2) gain more control over their own advertising destinies. The marketers chortled that they already spent so much indirectly by underwriting the cost of a $60 billion TV advertising marketplace, dismissing Marans' plea as just another attempt by an agency exec to go back to the well. Marans would nod his head in agreement, note that yes, marketers do indeed spend all that money on TV advertising--so wouldn't it be smart of them to spend it on a reasonably informed basis? Marans even had a direct and logical economic basis for his argument. If advertisers kicked in, say, $250 million worth of direct TV research spending, that figure would amount to just 0.4 percent of their total TV spending. So if the research only improved the effectiveness of TV advertising by 0.5 percent, it would pay for itself.

Marketers never bought into that pitch, and we fear it might be for even darker reasons than the fact that they're skinflints. We wonder whether they may have been afraid of what better TV research might have really shown them about the effectiveness of TV advertising. But that's another column.

David Marans has gotten out of that game. Gale Metzger is all but retired. And we're not sure where Rick Aurichio is. And frankly, we don't know where either Apollo or the portable people meter may be, either.

At least one player appears to be acting in earnest, trying to build not one, but two better mousetraps. Arbitron today disclosed that it has invested $52.7 million related to those two projects in the third quarter alone, up 12 percent from the same quarter last year. That tells us that Arbitron at least is committed to building a better future for audience measurement. The agenda of other stakeholders is not so clear.

We've already ranted on Infinity Broadcasting's decision to let go of its corporate research department the other week, and specifically, that it fired research guru Tony Jarvis. So we won't go into that one again, other than to remind you that Infinity's move was not a very good sign for support of the portable people meter.

As for Apollo, the godly single-source research system being championed by none other than Procter & Gamble, the picture is even murkier. It's clear that Arbitron is committed. But what about its partner VNU, which according to today's MediaDailyNews, has gone behind Arbitron's and the Apollo sponsors' backs, and has begun pitching clients on an alternative system that would provide much of the same utility as Apollo, but for a fraction of its cost. And here's where the confusion comes in. Don't get us wrong. We don't have anything against fusion per se. It's just another research technique--another potential tool to be used by researchers to do things they cannot do by other means. But no one should be fooled into thinking that it's a substitute.

To be sure, fusion research isn't necessarily a bad thing, and we've seen some very clever applications of it, including VNU's own collaboration with WPP's Kantar Media Research unit. The fusion of Nielsen's TV ratings data and Kantar's MARS pharmaceutical category research has provided some very valuable insights for drug marketers at a time when they need to understand how consumers really think and interact with media. But according to better research minds than ours, fusion research is only as good as the databases it fuses and the statistical techniques it uses to fuse them. In other words, it's a buyer-beware scenario.

Now we're sure that whatever VNU did to fuse its Nielsen and ACNielsen databases was done with the best statistical goals in mind, but it will never be a replacement for the real thing. And why VNU would bring it to market now, even as it claims to be working hard to convince marketers to commit more deeply to Apollo, we can only fathom.

It could be that one part of VNU isn't talking to another part of VNU, but we find that one hard to believe. A company like that doesn't do anything by accident.