Kelvin MacKenzie may just go down in history as the man who revolutionised the UK's media research industry. Not in the way he currently hopes or expects to revolutionise the media research industry, perhaps, because this isn't just about radio. MacKenzie's primary goal in his wrestling match with the radio establishment is to counter the injustice heaped on his talkSPORT station by the current radio industry currency, Rajar.
MacKenzie knows that Rajar is fundamentally flawed because when talkSPORT has some plum sporting rights on an exclusive basis, its ratings aren't affected in the slightest - or certainly not in line with the expectations of other pieces of research and anecdotal evidence. Rajar's accuracy is akin to the sort you'd get in surveying the ocean floor using a lead weight and a fishing line.
So MacKenzie has been pushing for the use of new electronic metering systems as opposed to the self-completion diary system currently used by Rajar. He's achieved surprising progress despite alienating many potential allies with his old-fashioned approach to diplomacy and the art of winning friends and influencing people.
But the most interesting thing to emerge is the notion that if (and at this stage it remains a very big "if") the new metering techniques can be made to work for radio, they can be made to work for TV too. And, at a stretch, posters, newspaper and magazines. In other words, one device can, in theory, measure a person's entire media consumption. Much of the running on this one has been made by the Radio Advertising Bureau, which has a long track record in investigating the effects of various media in combination with radio. But surely we shouldn't get too excited about the potential for single source data quite yet. "Why not?" Justin Sampson, the RAB's managing director, asks. "Both the technologies being assessed by Rajar can measure other media than radio. The technology exists already by which different sectors can come together to provide the numbers that people trade on. That in turn would lead to economies of scale, which would free up more resources to look at the qualitative impact of various media too. You could also more accurately assess the effectiveness of various media used in combination."
But this is surely all blue sky stuff at this stage? After all, we're still a long way from actually getting new technology in radio research.
Sampson says: "You've got to bear in mind that research contracts are often on a long-term basis - although Rajar is a rolling contract, the Barb contract is basically to 2006. It will be difficult to imagine change if we don't start talking about it now."
And the RAB isn't the only one already talking about it. The Barb organisation and its media owner backers are well aware of potential developments in this area. They are mildly sceptical about the level of demand for such a product - or whether the technology could really deliver something that the industry would be happy with.
Others too are up to speed. The Media Futures Group of the IPA recently asked one of its members, the PHD managing director, Morag Blazey, to canvass member agencies on this topic. Today's planning techniques are absolutely crying out for research that would reveal a more three-dimensional understanding of the behaviour of consumers in the round. "The goal is for research that is 'consumer in' rather than 'media out'. We're asking agencies to think about the prospects for that and what they would require," Blazey says. But she admits we're probably a million miles away from anything concrete in terms of technologies and methodologies.
Or are we? Not if you listen to George Michaelides, a founding partner of Michaelides & Bednash. You can already do it yourself if you want to.
He comments: "Existing research is knackered, I think everybody really knows that. Commodity industry data seriously undervalues too many media and the data doesn't tell you much anyway. Media owners should always do their own research."
Michaelides & Bednash is a pioneer of an approach to media strategy that sees no real need for any quantitative data whatsoever. Having identified a target market - on an attitudinal basis rather than on crude demographics - the agency then conducts focus group research to get a picture of the group's media preferences and its attitudes to various media properties. It is then in a position to assess the value of various media opportunities against the client's communications plan. Whether this notional value comes out at a discount or a premium to Barb-derived station average price (in the case of a TV opportunity) is seen as irrelevant.
"I think Kelvin MacKenzie is absolutely right to go ahead. Good luck to him. When media owners do their own research, they usually come up with something that makes you think and maybe want to follow it up with research of your own," Michaelides adds.
But if the M&B approach works, why aren't more advertisers using it?
Michaelides says there's an in-built inertia at work. It's easier, in effect, for a client brand or media manager to do nothing. They are often happy with a continuity with the past and established ways of doing things even though they are aware that those ways are deeply flawed. And if advertisers do nothing, mainstream agencies won't be inclined to drive this forward either.
Rival agencies say that's nonsense - clients need massive amounts of quantitative data that they can assess on an absolute basis. And if they sometimes seem reactionary, then maybe that's because their demands are so taxing. But they don't necessarily dismiss the notion of single source data out of hand. Bernard Balderston, the associate director of media at Procter & Gamble, certainly doesn't. "Clearly this idea has been something that a lot of researchers have thought a great deal about. The reality will depend on what the quality of the data is and the degree to which it can be applied to practical decision-making."
Although there have been several clever pieces of pan-media research that basically manipulate existing data, there has never been a thoroughly reliable method of evaluating a multimedia schedule. And advertisers would love to be able to do that. But, Balderston argues, any change should not be made at the expense of the high degree of resolution currently available in existing joint industry research. "My reservations where single source multimedia research is concerned would be about whether it would give you the kind of detail you get from NRS and Barb. Barb especially offers you all sorts of bells and whistles. I think it would take a lot to convince the industry to move back from that."