That much quoted complaint, "I know half of my advertising is wasted, I just don't know which half" (attributed alternatively to Lord Leverhulme or John Wanamaker, depending on which side of the Pond you live), will forever be laid to rest if the assorted powers behind Project Apollo, a joint US initiative from Nielsen and Arbitron, have their way.
Apollo, which combines data collected from consumers using Arbitron's Portable People Meters, with purchase data culled from Nielsen Homescan consumer panel to form a kind of "single source" measurement tool, is backed by the combined might of some of America's bluest of blue-chip advertisers, including Procter & Gamble, Unilever, Kraft, Pepsi and SC Johnson.
Mike Hess, the OMD global research director, and a member of the project's steering committee, says: "Apollo lets you know not just which half of your advertising works, but against which people, so that's an added refinement. Everyone is excited about the concept."
Don Gloeckler, the Apollo charter member and the P&G manager of media research, north America, says it's better than anything else out there because of the multimedia aspect. "Because it links that media exposure information to household purchase information, we can expand the consumer groups we're targeting with our messages beyond demographics like age and gender, and build in household purchasing."
According to Brad Agate, the senior vice-president of research at Horizon Media: "It's really an attempt for mainstream media to catch up with multimedia."
But the project, three years in the making, has had its glitches. Not least is the amount of time it's taking to develop. "I went to the very first Apollo meeting almost three years ago," Hess says. "But the key people on the panel wouldn't be willing to waste three years if we didn't think that there was a lot of promise here."
Thom Mocarsky, the senior vice-president, press and investor relations for Arbitron, admits it's taken longer than anticipated. "But we're very pleased with our seven pilot subscribers," he says. "What's great is that it's collaborative. The industry is working together to help us build and design the service, and we're helping them build their business case."
Then there's the cost factor. Arbitron has reportedly incurred losses of $1.8 billion on its half of the project. And no-one's quite worked out how much to charge potential clients yet. "This is an expensive system," Hess says. "So the question is, are we getting enough out of it to make it worthwhile?"
With Apollo initially asking for 0.5 per cent of clients' marketing budgets (on which they later backed down), Mocarsky agrees it's not inexpensive. "It's going to cost a fraction of a per cent of clients' national adspend," he says. "The seven subscribers we have spend about $7 billion on advertising in the US, and even a fraction of a per cent adds up to real dollars."
Agate says clients that stand to do well are those that produce packaged goods and do a lot of broadcast advertising. "This is a huge undertaking," he says. "Cars and big ticket items might be more difficult to measure, and print ads require another layer of data."
But it is still a pilot programme. "With 5,000 households and 11,000 people, it's not big enough to have the kind of granular detail a mid-level brand would need," Mocarsky says. "So the purpose of the pilot project is to concentrate on big volume advertisers."
And just why did the big volume advertiser P&G get involved? "It helps us become better marketers." Gloeckler replies. "If you think about the data that's available today, it's often used primarily as a currency to buy and sell media. Apollo is supposed to be more a marketing panel not a currency, and it's designed to help marketers understand their consumers. And that broad consumer understanding drives a company like P&G."
Assuming, of course, the technical issues can be worked out. "My understanding is that there are three electronic bands collecting information for Apollo," Hess says. "But the system is only picking up about 60 per cent of the commercials, which is probably too low. For it to work, it needs to pick up around 90 per cent. So you can't feel as confident that you're going to be able to draw these relationships between sales." Another problem is that Apollo has been slow to deliver data on multimedia including print, radio, outdoor and the internet, with some of the first reports only just starting to come in.
Eventually, it will be rolled out internationally, but for now, the focus is on the US. As for what rival operations are doing, Agate says: "There's a lot of granular data but no one can measure product usage. But who knows. Google is collecting this kind of data all the time, though there are issues of privacy."
Arbitron tried something similar when it launched its Scan America project in the 80s. But it proved to be a little ahead of its time, and folded in the early 90s. Meanwhile, Project Apollo is to roll out its next stage involving 15,000 households into the first quarter of 2008. "We hope that by 2008 everyone will have the information they need to make a decision. We believe the reaction will be a positive one and they think it makes business sense to pursue," Mocarsky says.