The techniques used to measure radio audience figures in this country are in urgent need of reform. At least that's what Kelvin MacKenzie thinks, and it's a view that he's not been shy in sharing over the past two years.
Those who have dismissed MacKenzie in the past should be warned - he is now armed with far more than just a celebrity profile and a talent for PR. Earlier this week the chief executive of the Wireless Group unveiled the results of a research project, which he hopes will prove his case once and for all.
The existing system used by Rajar to measure radio audiences draws upon 2,600 weekly diaries in which willing participants jot down their daily listening habits, breaking them down into 15-minute segments.
However, the Wireless Group's survey focused instead upon the potential of electronic measurement. More than 670 residents in the Slough and Bolton areas were asked to wear an electronic wristwatch that recorded the radio stations they listened to over a period of a week - 215 of these also filled in diaries for the purpose of comparison.
The results revealed that respondents listened to double the number of radio stations that their diaries recorded but for half the time (12 hours a week instead of a diary average of 23.2). What's more, 30 per cent of the periods of uninterrupted listening recorded by the watches lasted less than five minutes - significantly, Rajar only acknowledges people who listen for five minutes or more.
But MacKenzie's heart will be most warmed by the fact that of the four stations who doubled their reach in the survey, two of them - the Wireless Group's talkSPORT and BBC Radio 4 - are speech stations.
While it is important not to get too carried away with a survey of this size, few could argue that its results aren't thought provoking. But what impact will they actually have upon the drive towards a new system?
MacKenzie believes that the survey's results seriously undermine Rajar's credibility. "One of the most significant findings is that 30 per cent of all listening takes place in periods of less than five minutes," he explains. "This means that Rajar must be completely wrong because it doesn't take account of anything under five minutes.
"Apparently Rajar started this because nobody remembers less than five minutes, but now we have technologies that can do all this stuff for us.
Rajar also grosses up each five-minute listening period to equal a quarter of an hour. This is a ridiculous idea and if it was applied to television then all the audience figures would treble, I should think."
MacKenzie has been criticised for waging too public a crusade against Rajar. A recent Radio Audience Research report by Zenith Media urged him to "lobby behind the scenes with agencies and clients, rather than continuing to undermine the industry". But now that he has research to back up his rhetoric, does he feel that he's finally got the industry's backing?
"I think there is a change of sentiment," he says. "I think the industry has moved towards us. Behind closed doors, I think that everybody accepts that real-time research is coming. Rajar is operating a deliberate policy of slowing this down. It is not in the interests of its commercial shareholders to have a new research methodology in the market.
"Although Rajar is conducting research into new measurement technology, it has told me that it has no intention of publishing the results. The reason for this is that the results will look completely different from the existing Rajar measurements and on that basis the game will be up."
MacKenzie is referring to research currently being undertaken by Rajar into the commercial and practical viability of two electronic measuring devices. The first is the wristwatch that the Wireless Group is throwing its weight behind. The second is a pager that picks up an identifier signal in all stations' broadcasts.
So with all this research in progress, surely a new system is just around the corner? Justin Sampson, the managing director of the Radio Advertising Bureau, advises caution. He feels that there's little point rushing to adopt a new system if there's a risk that it will be quickly rendered redundant by changing listening technology. "Rajar has to pick an electronic measurement system that will cope with the future and the reality is that the future is increasingly digitised, increasingly fragmented. For radio alone you can listen to three or four different distribution platforms - internet, analogue, digital and satellite TV. Rajar cannot jump on to the first horse it feels like riding."
But what of the results themselves? John Billett, the chairman of Billett Media Consulting, feels that if the Wireless Group's findings are accurate, they will change the way advertisers see the medium. "What we have got here is a really big commercial issue. Here is what could be a better technique, which shows overall there are fewer listening occasions. That will mean that the medium is fundamentally less attractive to advertisers," he says.
"The national and speech radio stations have done better than people expected. What that says is that the medium is capable of far greater diversity than we thought. If this research is right, it has within it all of the evidence to create new forms of radio that will become more attractive. This changes the definition of listening, it changes the emphasis from frequency to reach, it changes the emphasis from music and talk, and it opens up fantastic opportunities."
A more diverse medium able to reach a greater variety of people will enable advertisers to exercise far greater precision in their radio marketing.
But what do advertisers really think of this debate?
One company that has enjoyed great critical success recently for its Hamlet cigar radio executions is Gallaher Group. The restriction on advertising for tobacco products has given radio a renewed importance to brands such as Hamlet. David Cook, the brands marketing manager at Gallaher Group, is not losing sleep over the issue of audience measurement: "In regards to the finite principles of whether the audience figures of Jazz FM are up X or Y, it's not something that we spend a massive amount of time debating," he reveals. "We trust the judgment of the framework within which audience figures are measured and will continue to do so unless someone is able to give a different and better framework. From a client point of view, we're all subject to a frame of reference in terms of choosing audience figures and, whether they're reliable or not, these are the figures we're given."