Kelvin finds Talksport's lost listeners

The Wireless Group's Kelvin MacKenzie isn't a man to take anything lying down. And when it comes to the question of how to measure audience figures for his radio station Talksport, the former editor of The Sun certainly isn't laid-back.

He has been embroiled in a row with the incumbent measuring agency, Rajar (Radio Joint Audio Research). He says its "diary" system, by which a sample group of listeners write down the stations they tune into, is inaccurate because it relies on memory.

Mr MacKenzie advocates a system where sample group members are given a wristwatch that digitally tapes their "audio experience" every four minutes, recording snapshots of what they hear. These can then be analysed, noting any radio stations listened to.

Initial tests show that stations like Capital Gold, XFM, Classic FM and Jazz FM have fewer listeners than indicated by Rajar's research method. But Mr MacKenzie's Talksport fares better: its audience jumps from 2.4 million to 8 million. Radio 4 and Radio 5 also benefit, he claims.

"Why should I sell [the advertising for] my audience at a 75 per cent discount?" he argues. "I was thinking of suing Rajar for costing me a load of money. I'm not putting up with it any more."

In September, Mr MacKenzie is planning to launch his own system for measuring audience figures, with the slogan "It's time to tell the truth".

The listening figures for the London region will be published monthly, which is bound to infuriate the stations on the other side of the argument.

Rajar claims it is researching other systems, and is due to report next spring on improvements to its current method. "There are pros and cons on both [sides of the argument]," says Jane O'Hara, managing director of Rajar. "My view is there will probably be a hybrid system: diaries for some stations, watches for others."

Ms O'Hara argues that the wristwatch method is expensive. Stations already pay Rajar £4.5m a year for its service but, she says, this cost could be increased by as much as sixfold if sample groups have to be provided with watches.

Such claims get Mr MacKenzie's back up. "It is absolutely untrue. Simply wholly untrue."

But one radio executive believes small local stations could lose out if Mr MacKenzie gets his way. "He is saying all that matters to him is what gives better results for his radio station, even if it means spending three to four times more," says Phil Riley, head of radio at multimedia company Chrysalis. "It could potentially rule out the little guys from having an audience measurement system."